ElevenLabs Masterclass: AI Voice, Audio, Video & Agents

1. Welcome: Welcome to the ElevenLabs AI Master Class. AI-generated voices have become incredibly popular and capable, but ElevenLabs can do so much more than just text-to-speech. You can design and clone voices, create sound effects on music, produce audiobooks and podcasts, dub content in other languages, generate visual content, and even build AI agents that can interact with users. In this course, I'll show you how to use these tools through practical examples and show you exactly what ElevenLabs is capable of. You can understand not only how ElevenLabs works, but how you could apply it and use it for your own professional work. My name is Jose Kuchii and I'm a visual designer with over five years of experience working with AI tools, creative content, and digital workflows. I have spent a lot of time working with these AI tools to figure out the best way to integrate them into real projects. Throughout this course, I'll share with you the techniques, workflows, and tools that have helped me along the way. We will start with the fundamentals. You will get familiar with the interface, the accounts, the pricing, how you can set it up the best way, and then we're going to move into prompting, which is essentially how you communicate with this tool. Once we have those under our belt, we're going to move into all the tools that ElevenLabs has for us. You will learn how to find and create your own voices, how you can transform existing ones, how you can dub audio for podcasts or audiobook, how you can create sound effects and music, and also how to make visual content with ElevenLabs. Along the way, we will also explore studio and other production tools for larger projects. Once you're comfortable with the tools, we're going to move into some practical projects. You will work with assets and templates, image and video generation, lip syncing, upscaling and flows to see how these tools can be integrated into a much larger creative process. And finally, we're going to explore 11 agents. Over there, you're going to learn what they are, how you can use them, and we're also going to create one together where we test and deploy them. By the end of this course, you will have a broader understanding of 11 labs, but more importantly, you're going to have had hands on experience creating those projects. So let's go ahead and get started. 2. What is ElevenLabs: Before we get into how to use ElevenLabs, let's first talk about what it is and what are the use cases. So ElevenLabs is actually a AI company and they specialize in generative audio. Generative audio basically means when you generate audio, and given that this is an AI tool, you're making audio with AI. Now, previously, there has been a lot of technologies regarding generative audio, such as text-to-speech, speech-to-text, and other stuff. However, those rarely involved AI recently with this implementation, the workflow has become a lot easier and a lot more accurate. When you type something and you use punctuation, instead of it sounding super robot, now with AI, it sounds a lot more realistic and the other way around speech-to-text, it can catch onto your words a lot better compared to the previous technology that was out there. This right now is the homepage of ElevenLabs. We will talk about account and pricing in a little bit. But just to walk you through what you can do with this tool, obviously, you can make a lot of instant stuff such as text-to-speech and then speech-to-text, but you can generate audios for your audio book. You can make podcast with it. You can maybe just talk to an AI for YouTube video or can even generate your own sound effects. There's a lot you can do here and this is a very popular tool if you're into making content or have some channel where you would want to post these contents or perhaps even monetize. A lot of room for experimenting with this tool, you can generate a ton of things, see which one works best and then fully commit with a longer audio clip. The way it works is that you have your account, you get a certain amount of credits and per generation, you're going to use a few credits depending on how long your audio is. If you've used AI tools before, the workflow is relatively the same. You give it a prompt, you give it an input, choose your model, and then you get an output. That we know what the platform is, briefly, just a general overview, we're going to talk about the account and pricing in the next lesson. 3. Account and Pricing: Before we get into how to use ElevenLabs, let's first talk about what it is and what are the use cases. So ElevenLabs is actually a AI company and they specialize in generative audio. Generative audio basically means when you generate audio, and given that this is an AI tool, you're making audio with AI. Now, previously, there has been a lot of technologies regarding generative audio, such as text-to-speech, speech-to-text, and other stuff. However, those rarely involved AI, recently with this implementation, the workflow has become a lot easier and a lot more accurate. When you type something and you use punctuation, instead of it sounding super robotic, now with AI, it sounds a lot more realistic and the other way around speech-to-text, it can catch onto your words a lot better compared to the previous technology that was out there. This right now is the homepage of ElevenLabs. We will talk about account and pricing in a little bit. But just to walk you through what you can do with this tool, obviously, you can make a lot of instant stuff such as text-to-speech and then speech-to-text, but you can generate audios for your audio book. You can make podcast with it. You can maybe just talk to an AI for YouTube video or can even generate your own sound effects. There's a lot you can do here and this is a very popular tool if you're into making content or have some channel where you would want to post these contents or perhaps even monetize. A lot of room for experimenting with this tool, you can generate a ton of things, see which one works best and then fully commit with a longer audio clip. The way it works is that you have your account, you get a certain amount of credits, and per generation, you're going to use a few credits depending on how long your audio is. If you've used AI tools before, the workflow is relatively the same. You give it a prompt, you give it an input, choose your model, and then you get an output. That we know what the platform is, briefly, just a general overview, we're going to talk about the account and pricing in the next lesson. 4. AI Ethics Overview: With all of these new AI tools coming out almost every month now, it's really important for you to brush up on the ethics of using AI if you're not already familiar with them. If you're a beginner using 11 labs or any other AI tools, you may be wondering why sometimes it's considered unethical. Even if you know and you're still not convinced, it's important for you to realize the risk and why it could potentially harm other people. For this lesson, we're going to talk a bit about the ethics of using AI tools, whether it's Chau BT, runway, M journey, or even 11 labs. There is a boom in AI tools nowadays, and it's important for you as the user to know about the risk and why you need to be using this tool, these really cool tools in a more responsible way. A lot of times you come to these AI tools to either make your work easier, help you with a project, or just have some fun. Those are all okay as long as you're not using it to harm anybody. Now, what I mean by harm? A lot of times these AI tools are being used to impersonate people. The inputs that are used to train these models are often taken from people who did not consent to it. For example, with a lot of art tools, there are many artists claiming that their works have been used to train these models when they did not consent. What happens when that artist who worked really hard gives their art it's being used to train these models and then you make something with it and then sell it. You right there have knowingly or unknowingly have contributed to this bad cycle that is harming these artists. Now with 11 labs, even though it's a great tool, it's important for you to take some time to read about how these models are being trained and how you could just use it for something that will not harm people. Give you some examples with 11 labs in particular, but this applies to any other AI tool that's out there. You should never be using these audios to impersonate real human beings. There are a lot of rules and regulations coming out every day about how to basically rules that prevent the harm of other people. If you generate a podcast fully using AI, you do need to label it as AI. If you use the script from the Internet that was also made with AI, maybe Chat GBD, it is your responsibility as a user to inform your audience that you're in no way trying to impersonate someone else or trick people into thinking that what you just generated is a person. You're going to be using 11 labs for your job or for your school, bear in mind that each company and institution has their own set of rules and regulations regarding it. So make sure you do your research and you don't put yourself at risk when using these tools. Now, for the purposes of this course, I'm just showing you how to use it. Use it to make podcasts, audio books, and stuff like that. These are just meant for you to explore and learn. If you do want to commercialize any of your works that you make by the end of this lesson, please take the time to read the rules and regulation of all the platforms where you would be uploading these products too, and again, label your work as AI. Because what happens is that when you submit your work in the same space as other human creators do, you will be flagged and according depending on which platform you're doing it on, there will be consequences. So be responsible with your generations. Don't use it to harm or impersonate anyone, label all your work. And if you choose to make money, monetize your work with either 11 labs or any other out there, do it in the proper fashion. Go to the platform, see what their guide is. Almost every platform has it, whether it's on Instagram, Bhand, Adobe, wherever you want to do it, go and navigate on their website, read about it, and follow their steps. Now you know a little bit more about the risk and the potential harm that you might be doing if you were to use AI in the wrong way. We can now move on and continue with our course where we learn more about this particular tool. So be careful out there and let's go. Oh 5. Prompting Basics: In this lesson, we're going to go over how to prompt with ElevenLabs. So a prompt is the directions that you give an AI tool to follow. And in this case, these are the words that you want 11 laps to execute. Whether it's a voice over, a text-to-speech, voice changer, or any of the other tools that we're going to take a look at later, there is a general style that you need to follow, and it's actually pretty easy compared to the other AI tools out. Head over to text-to-speech where we're going to practice with the different prompts. Now, don't mind the tools on the right side. I will get into what this is in a different lesson. But first, let's go ahead and do a simple practice. We're going to start typing in something. I'm going to have ChatPT give us a very simple story and then we're going to convert it into a decent audio. Give me a four sentence story about a dog. Just very simple. To make this better, let's do something like this. Make a podcast episode based on this with only four sentences for dialogue. All right, so let's copy the podcast part. The reason why I chose podcast is because this way we can play around with the different emotions. I'm just going to paste this here. Feel free to write your own text. But let's get rid of all of these tags and just put in the text that they say. Um, I'll put a paragraph break after each sentence. So let's first start by choosing a voice. For me, by default, it's George. The voice is the model that is reading your text. So if I just hit George, I'm not gonna change anything here. Let's actually reset the values just to make sure that we're all using the same settings. Let's hit Generate speech. Every morning, a scruffy little dog named Max wandered the quiet streets, searching for scraps and friendly faces. One rainy afternoon, Max huddled beneath a cafe awning, shivering until a kind barista spotted him and stepped outside. Hey, there, buddy. You hungry. From that day on, Max came back until one morning, the barista opened the door with a smile and said, want to stay. That's just the model reading our text. It's very monotone, it's not exactly exciting or dull, it's just very flat. So here are some things that already are included in this text that partake in that style that I was talking about earlier. Every time you add a period, that's an indication for the model to pause. Then when we have a, there's three dots, it's going to be a longer pause. If you do this, this is going to be a different cut. We're going to incorporate these within our prompt going forward. But right now we have dashes, we have the three dots, just a period, we have a comma. Those things are already included. And when I played the audio, you saw how it paused. When the sentence ended, it again paused here between the two words and it read it just fine. Now we're going to talk about how we can add emotions because we have a Brista here and then we have just a narrator. When you use different adjectives, ElevenLabs is going to understand that as the emotion it needs to execute. We can say, let's see. Right here when there's dialogue, we can say set the excited Brista let's see. Set the Brista in a excited way. I'm using the word excited here and then let's see what else can we add? Instead of this, I'm going to do an exclamation point and then maybe two question marks. So just this small change is going to make a big difference. Let's generate this speech one more time. Every morning, a scruffy little dog named Max wandered the quiet streets searching for scraps and friendly faces. One rainy afternoon, Max huddled beneath a cafe awning, shivering until a kind barista spotted him and stepped outside. Hey, there buddy. You hungry? Said the barista in excited way from that day on. So you saw how it says how the George here said that, Hey, they're buddy, you hungry. So that was the difference between adding two different signs versus just having one. Now, if you go to if I want to take this a step further, you can add pauses and all of that, too, which we'll get into later because there's a tool within studio where you just click a button, but you can add Pause is here, like three dots, and then maybe another one here. Now let's see how that works. One thing that you can do is when you combine the two sentences, put them right after each other, it's not going to pause as long. Every morning, a scruffy little dog named Max wandered the quiet streets searching for scraps and friendly faces. One rainy afternoon, Max huddled beneath a cafe awning, shivering until a kind barista spotted him and stepped outside. Hey, there, buddy. You hungry? Said the barista in excited way? From that day on, Max. So that's the changes that we're seeing. And moving forward, the only prompting you would be doing is giving 11 labs a story line like this one. We're going to look at how we can add sound effects, longer pauses, and different sorts of audios with the other advanced tools. The traditional form of AI prompting where you describe or you command for something, for example, make me a sound effect of heavy rain within the forest. That's an example. That will only take place when you're making sound effects with 11 laps. But for the rest of the other tools and there's so many, this is the prompting that you're doing. You can either generate these story lines yourself, type it in here and as you saw, it's really easy. Or you can have another tool like ChaGPT make something for you in a matter of seconds. This is just a lesson for you to know what we mean when we say prompt. It's not your traditional prompt, where there's specific formulas to follow. Just keep in mind the punctuation. Each of them mean different things, as we saw with the commas, dashes, three dots, stuff like that, and be very generous with your adjectives because those can mean your AI voice model going from monotone to either excited, angry, SAD, whatever adjective you want to use. And we're going to be seeing that a lot in the further lessons. Without further ado, let's jump into our next chapter where we look at the mini ElevenLabs tools. 6. Voices Library : Let's take a quick look around this platform and see how we can right away start making voices with ElevenLabs. This is the page you will see the homepage after you've signed into your account. So right away, there are some latest from the library voices. Essentially, we have a library where people can add in new voices. Of course, there's a procedure that they have to go through and it's not low quality voices at all. But this is a library that ElevenLabs manages and they're releasing new voices every day. Let's give this one a listen, for example. AubukieraOsk Terra. That's in Hindi, and then we have, for example, a young woman Silas. A So this voice right now you can see it's being called female mature voice, and it sure did sound that way, calm and deep. So I just refresh my page to show you that every time you do it, you're going to get a new set of voices. So, for example, let's go on this. Evil is evil, Lesser, greater. Idling. It's all the same. So this is serious and grim, and you can see it definitely did hit that mark. Now, down here, you will notice it says English preview. If you click on this, you get to switch between languages. These are the current languages that are available. It's going to be the same voice, the same tone, and all, but it'll be in a different language. So let's go ahead and try Russian, for example. Chive Katorn D Ciba Nemo switching Devi Camibe. City er SchimaFantasia, solitaria, Poggi transformer completa Menchi HmunogunKab Hari Hunaja, JonikOtmaFatra and 40s Helps Dantwsen The man lipped. You can see how seamlessly it switches between these languages while maintaining the voice and the tone. If you speak any of these languages, you can see if it sounds realistic or not, but from what the community says, it does hit the mark pretty well. Okay, so if you wanted to explore more, you can go over here or simply go down to voices. This will take you to the voice library, and we have a lot of options to choose from. Right away, you will see that there are checkmarks on some of them and that just indicates that they are high quality and they've been used a lot. You can see the name of this person and if you just hover over it, you're going to see a little bit more information. The description of the voices can really help you choose the one that you. Example, David here is a news reader, and he has a clear and crisp voice. He's middle aged, professional American, and then you can just gauge on whether or not that's going to fit to your project. Let's go ahead and preview this voice. Just click on it. Later that same evening, Detective Carlson received an anonymous tip directing her to a second crime scene. What had at first seemed a random event. And then once again, you can switch between the languages. Now, you can see here that the language selection is different from the first voice, and that's because each voice has its own language selection. So here we're actually getting Arabic, Polish, Alien, I think those are the only ones that are different. We're getting Hindi, English and French like we did before. And down here, you can see the flags. So it comes in these two languages, and then five more. These two, two more, and it's just going to change as you go on. In the Explore page, you can search for either the name of that voice or the way it sounds. For example, I'm looking for a calm voice, and I'm getting a lot of narrative and story, informative, educational going to give this a listen. Here, we have a calm, well spoken narrator, full of intrigue and wonder for nature, science, mystery, and history with a smooth and velvety tone. You can see that definitely did sound calm and it's actually one of the adjectives that's being listed here. Right away without having to click on it or hover over it, I know exactly what voice I'm going to get. Now, you can take a look on the right side, which is the categories that they go into. If you open these in a new tab, let's just open one. We're going to get a library just for informative and educational. You can see that's the category that I went into. Close that guy. Right next to it, you can see two Y or zero D. That just means that how long this voice is going to stay if it were to be removed. If the owner of this voice removes it from the voice library, it will become unavailable for you after two years. But here, for example, it will be removed immediately. If you want to use this long term, definitely go for the ones that have a longer retention compared to something like this. Next, we can see how many users have used this voice. The first one by Adam Stone, you can see it's pretty popular. 438,000 people have used this voice, meaning that you've probably heard it at some point. The people who are crazy enough to think they can change the world are the ones who do. There we go. Then there's a plus button, which means that you can favor this voice for later use. Click on at once, and then it basically goes in this tab, which we'll get into later. Now you can see that it went from a plus to this icon. I can use this voice to either generate an audio to text-to-speech, speech-to-text, whatever that I have to do. Next is more actions. You can copy the voice link the voice ID, and then view similar. If you've used other AI tools, think of the voice ID as the ST number. When you have the ID and you're trying to generate your voice, it will keep it a lot more consistent. It won't functuate as you're putting in different adjectives, different tones. I will remain similar. But usually, if you use one of these pre made voices, you don't really have to use it. The link is just a link to that voice if you want to share it with someone and then we have similar, which gives you more voices that sound like Adam Stone. Now let's explore the things up here. This was our search bar. We were able to look for an adjective, maybe someone's name. Let's look for someone called John Doe. There we go. We have three options. You can also look with search with age or gender, maybe senior can do male or maybe teen male or something. Anything that we want. So for example, English teen youth. Hi there. I'm Archie. If you're after a young, energetic and authentic British teen voice, then Hey, I'm Ethan, and I would be a great choice if you're looking for a teenage voicedor. You can see how those search bars affect. You can see how searching for a certain thing does give you the correct voice selection. Some of these voices will be charged a bit more than the others. With the plan that I have, you can see these ones are okay for me to use. They're the popular ones, and there's no dollar sign over here. But for these, you can see it's $0.02 and that's only for a few of these voices. You can see most of them don't have that. Let's go back up here. This was our search bar. Right next to it, you will notice this icon. Which is basically when you get to upload a sample, close that up. You get to upload a sample and then 11 labs will find one that's similar to that. Let's say you have this voice that you want to replicate with AI, of course, be very mindful of copyright, but you can put it over here. For example, if you want a nice confident voice, for example, Michel Obama and you want to get something relatively close, can upload a snippet of her speeches over here to get something close. For example, if you're doing some animation and you have a character after Michel Obama, you can do that. Again, you really want to be mindful about how you're doing this. We will talk about AI ethics in a later lesson. But you can't just scrap any voice and try to get the exact same thing because that could lead to some big problems. Right next to that, we have a trending. If you click on it, there are a lot of popular voices as well as unique voices. What I mean by that is that the trending ones have been used a lot. These are the ones that we see right here. I'm going to click on this triangle just to go into the list again. You can see a large number of people have used it. If you want to be unique here, you shouldn't probably go for something like this and that's where this tab comes in handy. If I switch to latest, you can see this one, for example, it's not verified, but only 14 people have used it. This way, I'm being more unique when I have my voices in my videos and there's also options that are lower, seven people, five people, and the list just goes on two people. Getting left behind with AI feel like you can't keep up. You business needs a prompt engineer who matters. So you can see, even though they're new, some of them are not as high quality as the most popular ones. So it's a good idea to preview them, make sure it's something that works for you, and then proceed to use it. We have Latest, we have most users. So you can see this one is really high. I think this is a really nice way to just talk naturally together, you know, T plainly. Let's give it a shot. You can see it's conversational, it sounds really natural and that's perfect if you want to do a podcast or if you want to use your own voice and convert it to this tone. That's something really cool that 11 labs have, and we'll get into that in a separate lesson. But that's what most users will get you. Character usage basically means how many characters you get to convert to audio within your credit range. This is something we'll talk about later when I show you how to do text-to-speech. You will see exactly how many characters you're using. For some of the voices, it will take more credits while the others won't is what it's referring to. Don't mind this for now. We'll get into this later. At the same time, we have the filters tab, which lets you dig deeper and find what you really need. We have languages, first of all, a long range of languages. It's like Welsh, for example. You can even choose an accent if it applies. For example, it doesn't for Welsh, but let's see what would have an accent. Let's see if English has one. There we go. If you're choosing English, there's a lot of accents that you can explore. Let's say I want someone from Chicago, let's choose that. Next is the category, which is the tone that the voices sound and speaking. Think about what you want this voice for and simply choose your category. Going to do a very natural tone, conversational. Click on that. You can also click more than one category. I will do maybe something like this. Quality, any or high quality. Remember, high quality are the ones with the checkmark. Those may ask for more credits, but you can just click on any to get to have a bunch of options to look at. Next is gender. I think I'll do a female age. Let's do an old person. Notice period. These were the numbers on the side, how long you get to keep them once the owner deletes that voice. Let's say two years. Custom rates, include or exclude. This is for that character usage that we saw. Then live moderation enabled. This is going to moderate the type of text that you're giving to this voice. I'm going to hit Include for now. Let's apply the filters. Right now, you can see I don't have any and this will happen, but you get to create your own voices. So that's something we'll explore later. I'm going to remove a few of these just so I can get something to show you guys. Okay, so just a US Chicago accent. Exercise makes our heart stronger and our smiles brighter. That was one example. You can see it's a very recent audio. Only 32 people have used it. It's not verified, but it's still a voice that you can use. Let's try something else. We can do, again, English, try Canadian maybe. So here's a high quality audio. Met Gram, this handsome cat at our shelter. This boy is lovable. With his tabby stripes and cute face, he's sure to steal your heart in an instant. So that was one example. Let's try to find another one. Ladies and gentlemen, welcome to today's match. We have an exciting match ahead between EFC United and Team Impact. The atmosphere is electric. These are all Canadian accents in different categories. Next, we have create or clone a voice. This is that cool feature that I talked about where you get to either design an entirely new voice from a text prompt. You would say maybe an older gentleman with a thick accent, maybe a deep voice, calm, energetic, whatever you want to put. And it's going to ElevenLabs. We'll give you a bunch of options, samples. You get to build up on that and then make your own original audio. Now, the things that you create, you can also have other people use it. That's where all those new voices come from. Less than a minute, and we'll get into this in a different lesson. We have a voice clone option, so you can clone your own voice with only 10 seconds. I will speak into the microphone for 10 seconds and then 11 labs will study my voice. Then I get to use my voice with different text prompts. For example, if I don't want to record a audio book that's 3 hours long, I will just speak into the mic for 10 seconds and then paste the story line or just the story into the box and then 11 labs will generate that three hour audio for me. Next is the professional voice clone. This is not included in the plan that I have, but this is just for the most realistic one. But for most of you, since you are using this tool for the first time, these two should be enough. Up here, we have a feedback box, which you could just write your feedback. We have some other stuff up here. This was the Explore tab. My voices will have the voices that you favorited or made yourself. We have default voices. These are voices by ElevenLabs, so these aren't user generated. Just trust yourself, then you will know how to live. You can take this, use it like all the other voices, and just take a look at the many options. We also have collections, which you get to create when you are trying to organize your voices. For example, I could create a collection for my YouTube channel, one for my podcast, one for my audio books, and that way, everything will be a lot more organized. You're previewing a voice, we have the name here, the different languages it's available in Play button, you can skip back and forth, see how long it is over here. You can download the voice, and then you can just hide the player if you don't need it. That's the voices panel. If you click on this Plus button, it's going to bring you the same menu that came up when we clicked here. Now, in the next lesson, we're going to go over the playground tools that are located right underneath and see what else we can do with this tool. 7. Text to Speech: With text-to-speech, you can easily turn written text into genuine good quality audio. And this is the interface where you're going to do that. So this right here is the first tool available. Once you click on this, you're going to have this area, which is essentially your playground, and then you have all the properties, the different voice libraries, any sort of setting that you're going to need, it's going to be on the right. And down here, we have some quick templates for you to get started with. But of course, you don't need to use these unless you want to. This right here is where you get to write the text, and you're going to basically pretend to be the voice. So right now, I have Liam from the library, but you can choose anyone else here. We already went over the voice library. There's so many to choose from. And you can also use the voices that you have saved. Again, you can filter through any of these categories and find the perfect voice. But even if you have a voice that you think is perfect, you still have further ability to adjust that voice. For example, if I choose Adam right here, I can still come over change the stability, the panning, the language that he's going to speak, the output format, and everything else that's needed. So let's go ahead and see how this works exactly because there is some cool things that you could do with ElevenLabs in text-to-speech. So here you can type in the text that you want Adam to say, but at the same time, ElevenLabs has something called audio tags. And this is an example right here. So, for example, laughs, cries, screams, any sort of audio tag that you want to be implemented, you can add that in using the brackets. Can also use punctuation to work with the pauses, the full stops, and all that. And at the same time, we can have multiple speakers. So what we're going to do is make a very simple conversation between Adam and another speaker, and we're going to type in some sentences and use some of that audio text that we talked about. This way, you will get a better idea about how text-to-speech actually works and how you can use it to the full potential. So let's start with Adam. I'm going to have him you say, Hey, by using the exclamation point, it's going to be a little bit more excited. So Adam will be kind of energetic when he's saying that. And just to show you what that sounds like, I'm going to put another sentence after this. So, hey, how is it going? So, right now, I have used two punctuation marks, and we're going to see what this sounds like. I'm not going to change anything here, and we're simply just gonna click on Generate speech. Hey, how's it going? So you can see that he kind of stretched the hay and then it asked a question. Now, if I remove these guys and just not use anything at all, that's going to sound a little different. Hey, how's it going? So he did not pause, and it's less of a question now. So one way that you can control the voice is using those marks. So let's add in those question mark. And then we're going to add our second speaker. So let's make this someone else. I'll just go with Ellen. And now Ellen is going to do something else. So maybe she gets startled when Adam asks this question. So maybe yelps. Oh, my God. You scared me. Hey, how was it going? Yeah. Oh, my God. You scared me. So you can see how well that played out. And to show you the power of the audio tags, I'm going to remove this now. So now it's just, Oh, my God, you scared me with the exclamation points. Hey, how's it going? Oh, my God, you scared me. So you can see, even though I didn't put the audio tag, the voice is still pretty natural, and that's where I want to point to the models that we're using. So right now we are on 113. If you click on this, you get to see all the different models. But essentially, 11 V three is the most expressive. So the way that, you know, he comes in, I'll coal and, like, a deep voice, and then she comes in a little startled. That's because of our model. We do have multilingual models here as well, but right now, I'm just sticking to English. So if you do want to do this in a different language, I recommend moving to this one. There's at least 26 other languages that it supports. So if you see the language that you like, you can put it right here. And we have even more models. So you can see how right now it's saying that it recommends this model for Ellen. Sometimes it does that depending on the voice that you choose. There's also flash turbo and different turbo models. So it really depends on how much stability you want with the voices and how much expression you want with them. So these guys down here, they're usually good for quick conversions. Like, you don't want a lot of detail going into the voice generation. So you can see that it mentions conversational use cases. So maybe if you have a YouTube video where the person is just speaking, there isn't much going on. You can use a model like this. And we also have other models that it recommends us for to use it for, like, developer use cases. For example, if there's, like, AI chatbot on the website, you can use one of these models. But if you're going to build a story where there's a lot of audio tags, there's a lot of pauses, and motions, you want to go for the first model. So I'm going to stick to this guy, and I'm going to put back that audio tag. Just say what it is and then put it in brackets. It should turn pink. If it didn't either because it doesn't recognize that audio tag, like, it doesn't support that expression, or you may have put in an extra space, a comma, misspelled or other causes. But once it's pink, it's for sure and audio tag. You can add Adam back. You can see it picked up immediately, but you can add a third speaker if you want. So I'm going to do laughs. You're so easy to scare. Did you finish the homework? Let's add Alan. Again, you can see it picked up that there is a conversation. We'll do this size. No, da da dt. I was so busy yesterday. You can also use the audio tags between words, so it doesn't always have to be at the start of a sentence or at the end. So right here, we could do Snifs audio tag. They think I am coming down with something. And now we're going to generate. Hey, how's it going? Oh, my God. You scared me. You're so easy to scare. Did you finish the homework? No. I was so busy yesterday, but I think I am coming down with something. So there's our very simple conversation. But now we could go into each voice and further define the way the expressions are being made. So, for example, Adam right now is a little bit too intense. It's not really a conversational voice. So I'm just gonna click on Adam, but then go back to the page right here and make it a little bit more creative. So let's see what that sounds like. Hey, how's it going? Huh. Oh, my God. You scared me. You're so easy to scare. Did you finish the homework? No. I was so busy yesterday, but I think I am coming down with something. So now his voice is a little lighter. That's because I pulled it towards the creative side. Just be careful with this because when you're telling ElevenLabs to not be that consistent through the voice generation, the voice may kind of glitch if you're doing a very long paragraph or, like, a big story. But for a situation like this where it's, like, a sentence or two, works pretty well. Now when you change the models, you do get additional things that you can adjust. So this right here is for 11 V three, but I'm going to go to Ellen right now and switch this to the second model. You can see that it is recommended right here. Going to click on it. And now you can see that there is a lot more options. The only problem is now that we don't have that conversational layout with B three. We don't have that different speaker mode. Right now, everything that's written here is for Ellen only. So that's the downside. So you can kind of move around each model depending on what you need it for. But we can use Ellen right here. I'm just going to delete the lines that Adam had. And I'm going to, you know, play around with some of these sliders down here. So the first thing is speed. Is she talking really fast, really slow? You can adjust it right here. So, for me, I'm just going to lower it a bit. And with these sliders, you do want to be careful again about not going overboard, because if you do go all the way back or all the way forward, the voice may not be able to keep its consistency across, you know, different sentences. So whatever adjustment that you do try to keep it very minimal. So I was one here. I just came down to 96, so not like here. Next thing is stability. So do you want her to have different volumes at points or do you want her to be just monotone? You know, that's something you get to decide with this slider. So if I go all the way here, she's going to sound like a, you know, robot. But if I go all the way here, she may just be very hyper. So again, we can make very slight adjustments. The next thing is voice clarity. I'm going to keep this as it is. I think it's, you know, pretty clear. Next, we have style exaggeration. So do you want it to be, you know, the way she sounds right now, or you want her to be more excited? Do you want it to exaggerate all the different punctuation marks, audio tags, and all of that? Again, I'm going to leave mine as it is. And then, lastly, we have panning. So it's just about, like, the stereo field. You can think about like two speakers. Do you want it to be more towards the right or more towards the left? So just to show you what this sounds like, I'm going to bring this all the way to the right. And that way, it will only come out from one side of the speaker. We also have language overwrite. If you want to manually select the language that ElevenLabs produces for you, you can turn this on. So, you get to choose a language here. If you don't, it's going to detect what you have up here and then make that decision itself. Next is audio effects. These are additional files that you can add in here just to make the voice maybe more realistic, more suitable for a certain environment. These are all easy to add as well. For example, I could add in some noise, say she's in a cafe. I could filter something as well. I maybe she's on the phone. You can work with the space. This is the echo, and then we have the distance. So how far is she from the microphone? So I'm just going to do the first two first, and you can do a little preview before you waste any credits. Here's a preview of your voice with effects applied. Here's a preview of your voice with effects. So you can see that it only came out of the right side because of this. So let me just put this back. Here's a preview of your voice with effects applied. Here's a preview of your voice. And that's what it's sounding like. Lastly, is the output format. These are different formats you get to choose from. It depends on what you need this voice for. So you can see some of them are completely grade out because it's for a more professional plan. I'm currently on 11 Creative, so these two should be enough for me. I'm just going to keep it on the default, and, of course, we want the speech to be boosted. If you ever wanted to go back to what it was initially, you can click on Reset values. Let's generate the speech. Oh, my God, you scared me. No, I was so busy yesterday, but I think I am coming down with something. So now you can see that the voice is very different from the first model because we completely got rid of the audio tags, as well as a lot of the expression that was available in that first model. But I'm just gonna turn this loud, just so you guys can hear that cafe thing, and I'll remove the phone. I think that's making it harder to hear. So let's do this one more time. Here's a preview of your voice with Effects applied. Now we can better hear that cafe background noise. If you go onto a different model, you may get additional sliders. But again, only come down to these if you want something specific done. So this right here is the 11 turbo. It's pretty much the same sliders. So it really comes down to, like, the languages that you want to use or the speed of the generation. But V three is, you know, the best for conversations, and honestly, it sounds the most natural. You have an audio, you can come down here, you can play it, move forward, backward, give feedback if you want, share it with people, download it for yourself, and you can even hide the player. So if you download it, it's going to just go straight to your computer. It's going to be the same format that you chose right here. Now, if you're using V three, you can further enhance audio with audio tags, and that's something specific to this model only. That's how the text-to-speech feature works. It's really helpful, and the more that you play around with it, the more you're going to be able to find the perfect voice for your project and be able to keep that consistency across all of them. 8. Voice Changer: Now, while the voice changer seems like a very straightforward tool, there's a lot of things that you have to consider in order to get a decent and accurate voice conversion. This is the interface when I first click on it. We're still in the playground group. I just click on voice changer and this is what you should see. Essentially you input an audio, whether you record it now or have a pre recorded audio, and then you choose the voice and the settings, it converts it from your voice to whichever voice you choose. What we're going to do is record our audio directly into the computer and then just go from there. So click on record audio and when you do, you're going to get the option of which microphone you'd like to use for this. When you're ready, you can hit Start and you can see that we have a maximum of 5 minutes. I'm just going to read a piece from a short story maybe and then we can see how well it does over there. This is my original audio. Everyone looks up. A few people start jumping up and down, waving their arms in the air. Jake looks over at the towers surrounding the compound and notices the guards on full alert, training their rifles down into the compound where the men and women are making a commotion. So here I read things with a slower pace, but later I'll show you what you would get if you were to speak too fast or if your voice was just unclear. Now that I have my audio, I can either delete it to get another one or just proceed with the conversion. First, you have to choose your audio. I'm going to try both a female and a male so that you can see the difference. Let's go ahead and try female voice that is not similar to mine. Search for female. Just trust yourself. Then you will know how to live. She tore her gaze. Hi. Government of the people. There is no greater harm. Ideas are the beginning points of all fortunes. So here we have Dorothy with a British accent, which is an accent I don't have. So let's go ahead and click on this. This is the model that it chose for us, and I'm just going to leave it with the default setting. Now, one thing that I do want to point out here are these two options that you get to turn on or off. The first one is remove background noise. If you're not recording this directly into a microphone, you will need to turn that on. That way you will not focus so much on the cars passing by or someone hitting the table. And just focus on your voice alone. Now, speaker boos by default, it turns on, but I found that when it's turned on, it does distort your voice a little bit. I'm just going to run this with the default settings, which is what you see on the screen right now. Then later I'll turn it off and then show you the difference. Everyone looks up. A few people start jumping up and down, waving their arms in the air. Jake looks over at the tower surrounding the compound and noises the guard on full alert, training their rifles down into the compound where the men and women are making a commotion. You can see it struggled with some of the words and use a lot of s, which is something that I believe is caused by the speaker boost. Because I do have access to a professional microphone, I don't really need this one. So what I'm going to do is turn it off. And then reduce the similarity so that I could have more of British voice, I guess, something closer to what Dorothy sounded like. If you increase this all the way to high, it's going to really try to mimic your exact tone and pitch, and that could lead to further distortion. Let's reduce this to 40%. I will leave stability and exaggeration as they are. Now let's see the difference. Everyone looks up. A few people start jumping up and down, waving their arms in the air. Hake looks over at the towers surrounding the compound and noices the guards on full alert, triming their rifles down into the compound where the men and women are making a commercial. So that's still a little better, but we can go even lower. We can try for more stability. Give that another try. Everyone looks up. A few people start jumping up and down, waving their arms in the air. Jake looks over at the towers surrounding the compound and noices the guard on full alert, training their rifles down into the compound where the men and women are making a commercion. It's a lot better than what we had initially. It's still struggling with the accent part. Words like commotion, it does commercion an RN there. But that's something that we can fix with a different voice. But instead of going through all the voices, I'm just going to switch over to a male voice and try a different accent. Let's try something without an accent, like American male. With the dawn, fresh solutions light the path. To climb steep hills requires a slow pace at first. Hey, come here. Hi, you've reached Acme Corp. There comes a moment when the familiar path no longer feels like enough. A sunset is a day's way of saying goodbye, Grace. Okay, so I'm going to go with Liam over here. Let's click on his voice. So we have a new audio right now. Over and over, he hears the sound of the doors being slammed and bolted slammed and bolted. So now we're gonna try a male voice. God has given you one face, and you make yourself another. So there's George down there. We do have the option of different languages, but let's start with English. Over and over, he hears the sound of the doors being slammed and bolted, slammed and bolted. So it's interesting I said those two words twice. The first round, I didn't say well, but the second round it did. So there is some inconsistencies as there always is with AI. But for the most part, you can see that it's followed along the places where I changed my tone and the way I said the words. So it's pretty identical except for the fact that it's George. I'm going to do it one more time without the speaker boost and see if that's any different. Over and over, he hears the sound of the doors being slammed and bolted, slammed and bolted. Let's actually reduce the similarity to 20%. Over and over, he hears the sound of the doors being slammed and bolted, slammed and bolted. Was a lot better. I said the words right and that's probably because I reduced the similarity to 20% and removed these two options. What you can do with this audio is download it. You can also download it with the original name, delete it, and share it if you'd like. That's pretty much how the voice changer works. You could also try it with external voices, maybe do something in text-to-speech, and then you change your mind, you can just come to Voicehanger, upload what you got from here into here, switch to a different voice. Just keep in mind that text-to-speech is a lot more accurate than voice to voice. So keep that in mind if you're doing this for a more serious project. In the next lesson, we're going to go over to sound effects, which is little sound bites that you get to put in your video to either emphasize a certain action or to make something more dramatic. Let's move on to that lesson. 9. Sound Effects: Now, let's move on to sound effects, which is when you get to generate a sound effect with just a few words. And right away, you can see that there are some examples for you to explore. These are sample prompts. When you click on them, they'll show up into this box, which is where your prompt is going to go in and then you just hit generate. Let's do a car whizzing by. We can also try something else like a cat maybe. Down here, it's the duration. So how long do we want this sound effect to be? You can grab a slider and if you're not sure how many minutes or seconds you want it to be, you can just hit Auto. Then we have our prompt influence. How strict do we want the AI module to follow our prompt? If you wanted to have some creativity, you can keep it on the lower side. But if you want it to be exactly what you asked for, just bring the slider towards high. The default is 30, and that's what I'm going to leave it as. Over here, we have the history, so you can get different samples based on a prompt that you give it. Previously I did this one semiautomatic weapon, and I got four samples. Here you can press play to hear the difference. You can see all of them are weapons, but they start differently and they even sound different. Based on whichever what I'm trying to do, I can choose any of these samples and simply click on Download. What you can do is click on D plus to reuse the prompt and maybe you can add a new detail over here. These are organized by date. You can just have a look at what you've made so far and just keep track of what you're building. Is the duration and the influence. Okay, so that's about it with what you can do right away in the sound effects panel. We do have a Explore tab, just like the voices library where you get to view some of these sound effects that already exist. For example, let's do, let's see. A Lion. We can even do cinematic effects. This is urgent emergency broadcast. Attention resonates. A contentment breach has you can see all sorts of sound effects can be found here if you wanted to fine tune it. For example, this basketball thing is great, but maybe you want it to start with a whistle. All you have to do is click on use and it's going to come into your prompt box, and then you can add that detail. Maybe whistles blowing in the back. I added one more detail. Let's hit generate and see what we get. While that's happening, I'm going to just there we go. Let me replay the original for you so you can compare. And then we have four samples. All of them sound like a very intense basketball match. We could also in a similar fashion, remove a condition. Let's remove the whistle and crowd cheering and just do a silence basketball match. Let's hit Generate and I could just toggle that so that it's organized. If you have a lot of stuff in your history, there's also a search bar, that's really handy if you plan on generating a lot of sound effects here. So we can hear that basketball dribbling and there's the sneaker squeaking in the back. But as you saw, there is no more crowd, there's no more whistling. And that's because we removed those two parts from the prompt. That's it X. You can download any of these if you'd like. If you click on them, you can also just copy the prompt. Just like the voices library, there's a search bar. You can click on More to also look at it this way, most downloaded, most recent for something unique, and then trending audios. Up here, there's a bunch of categories to make search easier. Let's try something like UI elements. If you want to add a sound effect to your website, for example. Welcome to VCStom Graphics. Let's go back to Explore. There's one more thing that's really cool in the sound effects panel, and that is soundboard. If you click on it, it will take you to this whole different page. If you've worked with soundboards, this should look pretty familiar to you, but if not, don't worry about it. A soundboard is basically a board where you get to add different voices anytime you click on it. Think of these boxes as buttons and say you are having a podcast where you're reacting to something. Anytime you want there to be a drum, you would click on the drum button, and that's going to be recorded into your podcast. We have a bunch of sounds here already. Let's go ahead and try a few. I just went in the drum category. We can do some nature noises. You can see how I can stack them up as well and create that perfect cinematic sound effect for my video. There's also movie slash TV. These were your categories. You can also add a new row and not all of them have something in them. For example, movie slash TV doesn't have anything here. What you can do is click on them and generate one. Let's say I want to create one where people say boo. Let's say a crowd says boom in the satisfaction. Once I do that, I could give it a name. Let's call it boo. We get a bunch of samples. Then I could add any of these that I like to fill up this empty spot. Let's do number three. Now you can see there is Bo this empty button. You can save the preset, share it with someone, can also assign a device here if you have a keyboard or something that you can press keys on essentially. You can grant the permission and then have that connected to your computer. Instead of pressing these with your mouse or your cursor, you can click on the keys on a keyboard or a different device. You can also do is create presets. Right now, these are the defaults. But you can generate your own presets if there are certain noises that you need every time. For example, let's say I want to work with a lot of singular noises like someone saying something. I could start doing that. Just go to my preset, new preset, let's call it singular or just singulars, save the preset, and then you will get all these empty boxes. Click on any of them and start filling them in. A man saying, Hey, Hi. Hey. Hey. Hey. Hey. We got four generations. Pick the one that is most accurate. If it did not give you something you were looking for, you can just click on Generate again. I think I will add Hey. Hey. Hey. This one. You can see there's three versions, and that's fine because I could just cut it down. Maybe here we will do a woman gasping. Then here I will do maybe a car crash. Call it crash. Let's add this one car crash and traffic, maybe. I'm still on number three, so I get to replace it if I find something better. Sounds more like an explosion, but it's okay. And then number four, we can do sirens. Maybe police car, sirens. So there is the police car. Let's do one last one for traffic horns or honking. Traffic honking. So maybe the first one. So we have all of these sounds, and let's say I want to generate a car crash scene. All I have to do is press the buttons at certain times and make that perfect audio. So maybe we start with traffic, car crash, woman gasp, this, and then hey. Hey. Hey. So that was an example, but you can just play around with it. You can do That's just how it works. Then once you have something that you're happy with, you can even loop it. Turn on loop That's just how it works. This was the volume, if you ever want to play around with it. We also have the edit mode if you wanted to edit your sound. I'm just clicking on the man saying, Hey, and I could change my prompt and replace this. If you need, you can add more rows down here and just fill it up with whatever that you need for your project. Then up here we have the play button. We have the stop button or just clear whatever that's in here. This currently is not saved, so you can just save it into your preset. If you scroll down, there's a lot of explanation on how this works. We also have a bunch of soundboards for you to try. Let's try Goofy of, for example, and now we have a bunch of voices to try. G to hit loop and then lower the audio and start filling it in. What was that noise? That was a good so that's just how it works. You're just pressing a bunch of buttons to make your sound effect. Let's go back over here and close that up and then go back to generate. That was the sound effects panel. It's a lot to work with and you really have a lot of flexibility on how your sound effects sound. Now if you're not familiar on how to use these effects, basically, if you have a video and you want to emphasize a certain action or certain dialog, you can add the sound effects. For example, if there is a car passing by in your video, you can ask 11 laps to do a wish sound so that you don't have to actually get super close to a car and record that sound with your own microphone. It is also really helpful if you do any games or sketches as it saves you time by just not having you record sound effects from scratch, which can take a long time. For the most part, it's pretty accurate if you saw that you're not getting the sound that you want. Try going in the Explore tab, searching for something similar, and simply change it up to your liking. Now let's move on to voice isolator, which is the last tool within the playground category. 10. Voice Isolator: The voice isolator is another straightforward tool within 11 labs, and essentially what it does is that it removes the background noise from your audio and isolates your voice. So to test this, we're going to do one directly into the mic, and I already recorded an audio where I'm speaking next to the AC. So there is some noise in the background, and we're going to see how well it performs. This is the interface when you first come to this tool right over here. The only thing you can do here is upload an audio or record into your computer live. So let's go ahead and do one where I'm just recording my audio with you guys. I have a microphone with me, but I'm going to keep a distance from the mic itself, and then we'll compare. First, actually, let's do one where I'm speaking directly at a good distance, and then we'll move back and compare. That is my audio 17 seconds. I just read something off a piece of paper, and let's hear what it sounds like without any sort of isolation. Find a sister still living in her small apartment above a grocery store. The flat is crowded with people, friends who had fled the city and are now really tired. No one has appetite for food. Okay. What you can do is delete this record again, or simply click on Isolate Voice. They find a sister still living in her small apartment above a grocery store. The flat is crowded with people, friends who had fled the city and are now really tired. So that was with me speaking into a mic at a good distance. Let's just your ideal recording scenario. But what if I was standing next to a noisy environment? Maybe there's a fan turned on, and there's just a lot of white noise around me. So that's what I'm going to do next, and then we can compare how well this does. So I just recorded an audio where I have a fan in the background. They find a sister still living in her small apartment above a grocery store. The flat is crowded with people, friends who had fled the city and are now really tired. No one has appetite for any food. So it's really subtle in the background, but if I want to do this professionally, that subtlety is also what I want to get rid of. I just want my voice isolated so let's see how this performs. They find a sister still living in her small apartment above a grocery store. The flat is crowded with people, friends who had fled the city, and are now really tired. So notice how it completely removed that background noise. I'll play the original just for reference. They find a sister and then this one. They find a sister still this was only 15 seconds, but you can also do this for longer audios and it comes really in handy when you don't have a professional microphone, maybe you're recording in your phone. There's some stuff in the back and then you can just download it right over here. That's all you can do with voice isolator. You can isolate your voice. In the next lesson, we're going to move on to the products section of ElevenLabs. There's a lot of cool stuff there too, and there are a lot more in depth compared to what we're here. I hope you guys were following along and trying to mimic the demonstrations that I was doing in the lessons because in the next couple of months, things are going to get a little bit more advanced and feel free to pause or refer back to the older lessons because essentially what these products do is that they combine a bunch of these playground tools and they give you an option to make more complex audios. Let's go ahead and get started with our first product studio. 11. Studio Part 1: So studio allows you to create something from scratch. And right away, you can see up here we have some presets to follow, such as starting from scratch, which is you starting an audio that could go in any way, audiobook. So it will have tools designed for audiobook audio. We have podcasts and then we can just import from URL. Can access studio from the product category right over here. These are the options that could get you started right away. You can search for your projects right here. Right now, I don't have anything, so there's no results. But you can see there are some guides down here if you wanted to learn more about sound effects in game development, voice changer dubbing and all that stuff. First, let's start from the first option. Start from scratch. I click that, I think you saw that it made untitled project. Let me just go back and show you. This is my first project and they're just going to line up right over here. You can open the project, duplicate it or delete it, and when you click on it, you go into the project. This is how you can make audios from scratch and decide if you want to add sound effects, make it an audio book, make it a podcast, make it a narration, anything you want, really. First, let's go ahead and give our project a name. I will call this my scary story. Can start typing right over here. Then if we don't want to type right here, there's some stuff that we can add in like uploading a PDF. If you have essay you want to narrate, you have a short story written in the form of a PDF. This is where you upload them. We can add a video upload a video to add the voiceovers too. You can add a URL. These are some presets for you to get started with, and that's about it. Let's go ahead and try the stuff up here and then we'll come down for the other ways to add an input for this. Right over here, you can change your text into either different headings or just your regular body text. If you have a paragraph, you get to use this and show a history of what you've done. You got some undo buttons. We have a break. So when you add a break, the speaker is not going to speak. For example, if you have two paragraphs, you want to break between the paragraphs. You can add sound effects. Can even alter the pronunciation of our voice, which right now is George. We can direct the speech with our own voice, and then we can lock them, then finally get some information about what we're doing here. On the right side, we will choose our voices. If you hit Select voice, this is the same library that we saw in the previous lessons. You can search for it. Go to your saved audios, use professional voices only the default ones, what you used recently, and even your own voices. Let's go ahead and choose someone that whispers maybe. The moonlight dances across the room. I'm here to help you drift into deep, gentle sleep. You're safe. So I want to do a scary story. So I want a voice that can sort of whisper. So ASMR titles are kind of great for this. I'm just going to click on ASMR mic. Let's add it to our voices so we can easily access it later, and then we can just hit Apply. So there is this toolbar next to it, which is what we saw previously in I believe one of the other tools. So you still have a lot of control as to how the voices apply to your text. First let's go ahead and write something very basic for our scary story and see how we can get Mike to bring it to life. You can see right away as I start writing, it's assigning Mike to what I'm writing here. Then I could switch between George, maybe seven other people or just have one person saying this entire thing. So here's a little text that I just wrote. It's only three sentences long. Right over here, I get to change the layout for the text. For example, I could make one of them bigger, and maybe if I had two paragraphs, so let's just write one more. Following him as a long slender figure lingering in the shadows coming for his soul which he cherished once. So another paragraph, and then you can see how I can change the way the paragraphs look, and this could show that this is a really important opening and then maybe I would assign a command to it and do a bunch of other stuff. That's one of the stuff. If you hit Inter and hit a forward slash, you can actually add the commands. Instead of going from here, you can also do forward slash. Let's do a break. For 1 second, Mike is going to stop reading. You can also click on this and make it more. I do 1 second. Can do the same thing in between the words, forward slash. Let's do another break and I will make this one 3 seconds maybe 1 second. 0.1 is really little. Let's try 1.6. Then we can also do other stuff like forward slash, sound effect. Let's see where we can add a sound effect here. We can do right before on a rainy day, sound effect, hit Enter and then click on preview Raining let's make the duration like 1.4 seconds generate the preview. It's the same thing that we made in the sound effect generator. You get your previews, hit Auto if you're not sure, and then see which one you like best. So none of these are raining. It's still raining on a stone road with thunder in the background. Something more for 11 labs to work with. I regenerate. Let's describe the sound, and I will make the duration a little longer for, like, one sentence. So I like preview two better, hit a flive and there you go. So for this whole sentence, there's going to be this rain sound effect. If you change your mind about the sound effect being too long, just click on it and you get to mute it after a bit. So I'm going to make it a little lower and tear it. Yeah. So that's a lot better. I want it to be a background noise, not so much like the main focus. So that's our sound effect. We can also add the pronunciation editor if we wanted to add maybe a foreign word or something we want to make sure it's pronounced right. I'm just going to find a unique name that has to be pronounced. Unique name with right pronunciation, something random. I do want to test the accuracy of ElevenLabs. Let's pick up this name. Maybe N would be nice. Maybe 11 labs will read this as CN, but that's where we get to fix it with this guy. Can looks back and holds his breath. He knows what's coming next. Okay. So this is my story. Really short. Let's go ahead and have Mike start reading them. So he needs to read everything. 39 seconds. We can change the voice later. Down the road on a rainy day, there was a lone man walking back from work. His eyes were red from no sleep, his skin sunken from a lack of food. He drifts in and out of sleep but keeps on walking with his heavy body, for he knows what will happen once he stops. Following him was a long and slender figure lingering in the shadows coming for his soul, which he cherished once. Kian looks back and holds his breath. He knows what's coming next. So for me, it read key and write, but if for your case, it didn't work out, you have a specific word or different name, and ElevenLabs is just not picking up on it. You can go ahead and use the pronunciations editor. This is what it looks like. You either get to connect a dictionary which you uploaded or you made. And then you assign rules to this dictionary. For example, every time you see this letter, it needs to be pronounced in this manner. You can find your dictionaries here, you can add rules. Add new rule, input the word, the Alias, and then the output. Then if you don't want to use this dictionary ever again, you just disconnect it. You make a new one right here and these are the default voices that you get to get a preview with. If you're really into using this for foreign words or words that are hard to pronounce, I would highly recommend checking out ElevenLabs on tutorial on how to do it properly. You get to manage pronunciation dictionaries with programming. You have to install the SDK via your terminal using the Linux language. If you plan on doing this the right way, definitely check this out. Can find it in the docs from ElevenLabs. Normally it picks up on the pronunciations. For example, here it didn't read Can it read Ken, it should work for the most part. Let's go ahead and give this another regenerate now that we added more sound effects and this name. Once again, it's giving us free regeneration so that we're sure with our text before actually using our credits with Mike. Down the road, on a rainy day, it was a lone man walking back from work. His eyes were red from no sleep, his skin sunken from a lack of food. He drifts in and out of sleep but keeps on walking with his heavy body, for he knows what will happen once he stops. So that was my regeneration. Up here, we also have direct speech with your voice. For example, if it's reading a sentence in a really monotone way, this tool will help you use the voice to voice capability within ElevenLabs, something we looked at previously, and I get to guide 11 labs on how to read this sentence the way I wanted to. That was how we can use the tools, most of them for this one particular scary story. In the next part of this lesson, we're going to continue with the tools up here and then see how we can apply this to a document, a video, a URL, and expand our applications. I will see you guys in the next part of this lesson. 12. Studio Part 2: Now we're going to continue looking over the tools up here. So previously, we made this little short story. We had ASMR mice read it, added some sound effects and some pauses. But you can actually take this a step further. So say you have this sentence and you just don't like how either monotone or exaggerated the voice is reading it. What you can do is read it in the way you want and then have the voice copy that exact style. You can do that right over here with direct speech with your voice, right now, I just highlighted this sentence for reference, but you can do the entire thing or just one word, whichever you want. Once you have that selected, we just click on that and you can see it right here. What you can do is upload an audio or record your voice directly. Basically, let's first hear how Mike is reading this, and then we'll do our version. Let's regenerate the selection in case there's anything we missed. Down the road on a rainy day, it was a lone man walking back from work. His eyes were red from no sleep, his skin sunken from a lack of food. He drifts in and out of sleep but keeps on walking with his heavy body for he knows what will happen once he stops. Was a long and slender figure lingering in the shadows coming for his soul, which he cherished once. Can looks back and holds his breath. He knows what's coming next. That's what we have so far and interestingly, you read this as SN, but we can just direct it in a different way. For example, here, I wanted to pause between saying lingering. What I'm going to do is just cover this entire section, go over here and start reading this in the way that I want to. Choose your microphone and just start hit Start. Following him was a long and slender figure lingering in the shadows coming for his soul, which he cherished once. Kian looks back and holds his breath. He knows what's coming next. Once you're done, it's gonna show up like this, and you can either generate it or just start over for a different recording. I'm going to hit Generate, and we can now compare what we're getting here. So if you go right over following him was a long and slender figure lingering in the shadows coming for his soul, which he cherished once. Ken looks back and holds his breath. He knows what's coming next. So now that it was applied, let's hear what this new version sounds like. You can see on the side here, we have that conversion, and you can just click on it to see which part you converted. Following him was a long and slender figure lingering in the shadows coming for his soul, which he cherished once. Ken looks back and holds his breath. He knows what's coming next. And there we go. We were able to first of all, have this sentence read very slowly with the pause that we wanted. Also we had Mike read this as Kan and not seeing, which is what he was reading before. You can just really expand on this with some pronunciations if you would rather use this and this, especially for foreign languages, this will come in handy. Next tool allows you to lock paragraphs to prevent changes. For example, I don't want anything done to this top paragraph by accident because doing it, it's going to be a little bit annoying. All you really have to do is highlight that text and click on the lock. Now you can see a lock icon on the left side, that's about it. You can unlock it by grabbing it again and just clicking on the same icon. This one right here sends a report. Nothing too much you can do over there. You have your voices, change your settings if you need to whenever you make a change, just don't forget to hit Generate. You can also generate a section only by grabbing it and hitting. Use the bar down here to play a selection, change the speed, skip through, hit play your pause, and just see how many minutes or seconds your audio is. Can also zoom in and out. There's a timeline actually going to hit Zoom a little bit on this button first click Expand. You can see that when it starts, this is where the range starts playing, but I could just grab this and have it play at a different time. You can zoom in and out of the timeline like this. I believe it through so there's the pause. This is the for sound. I could just grab it somewhere like this and just make my changes in that way. This was with your classic text-to-speech feature. But what if you wanted to do this from a document? What I'm going to do now is just copy this and open a text document, nothing too fancy. Just paste it here, remove the different words and just leave the script. Once I have this, I'm just going to save it as a dot TxD. Call this my script. Then I will just delete all of this to get started with a document. Click on this and upload your document and see how ElevenLabs handles it. Can see we can upload a PDF, a EPAP, THD, HDL, or even a document. There's a lot of options. There we go. Now I have my script brought in from a document. If you're doing a audiobook, this is going to be really handy because you don't have to copy and paste anything. Then when we have let's try different voice here. Have Dorth read it. Now the name switched, go to hit generate, and we get to see it down here. Down the road, on a rainy day, there was a lone man walking back from work. His eyes were red from no sleep, his skin sunken from a lack of food. He drifts in and out of sleep but keeps on walking with his heavy body. Following him was a long and slender figure lingering in the shadows coming for his soul, which he cherished once. Kian looks back and holds his breath. Okay. So just like that, I imported text, it starts reading it. But what if you wanted to add something to a video? I have this video that I made with runway, which is another AI tool where you get to generate videos, and there is a voiceover currently on there, but I do want to try it with ElevenLabs. There's a lot of scenery in there. There's a lot of scevibes and it's a perfect opportunity for us to combine sound effects, breaks, voices and all of that in one project. So what I'm going to do is upload that video and I'll be back so we can continue. There is my uploaded video. You can preview it. It's a minute and 10 seconds long. Once again, we have a similar layout in terms of the voices and the tools up here. I I play this, you're not going to hear anything because there is no audio. But what we can do is go to the beginning. Let's open this guy right here and bring the video in the beginning. I think it should be shorter. Yeah, it's actually 53 seconds long. So what I'm going to do is just have this play and write a script for it. Let's start with a sentence. So that's our first sentence and based on how long it would take for it to become an audio, you can see the estimate down here. This is going to take about 5 seconds. If I had play, it's going to generate. There she was the lone survivor walking down the ruins. That's just an example. Obviously, we have to change the voice and do some other adjustments. So let's move on to our next scene, maybe somewhere around here. I wrote my script right here. I will add some pauses between each one just to make sure that every sentence is dedicated to one scene. This right here is going to be for the first scene. So right over here, we do need a break. Add a pause, 1 second. Then I'll add a pause after each sentence. Okay. And then we can add some sound effects. So here I think some crows in the back a little wind would be nice. Let's click on this right here and then go on the preview. So pros in the background. Let's try to keep it for the whole part where she's in this location. So both of these clips, once we're done, we can generate preview and decide which one would work for this scene. Okay. Now you can see that it goes over both of them. The next part is still about her being in this forest, so we can move on to the next one where she or she is outside the forest. This is when she actually goes in. So what we can do for both of these, actually, highlight this whole section and add another sound effect. So let's do Erie forest sounds with don't want to do crows again, maybe wind. I'm going to do another full ten second, and I think the first one is the best one. I added the pauses and the sound effects that I wanted. Just to walk you through, we had the first one with the crows in the background. There's a 1 second pause, the sentence, another second. I added a slow footstep sound effect in here because we're seeing there's something following her and that's about it. Now in terms of the voice, I do want to go for something a little different. Let's try searching for thriller maybe. Be it so dirty. No Atmos Segun Mosm Ignorovch. God has given you Let's do deep meal, something that would suit this horror look. Eso Hey, come here. I see that look. Our distrust Changing all things is sweet. Changing all things is sweet. Okay, I think I like Josh right now. So what we're gonna do is just grab all of this, choose Josh. Now you can see the name showing up. So she took to the woods looking for any signs of life. She found only memories of it. Okay, so I won't play you the entire thing. But essentially, this is how it would work. You have your video here. The timeline down below really helps you figure out what sort of sound effects you want, what sort of pauses, how long the pauses should be. And overall, you can create really good effects with this technology. So one thing that people like to do when they have a video like this, this video will be available for you guys to use, by the way. You can also put in, like, a vlog or something. Just focus on getting the sound effects right and then go ahead and add some music maybe or a voiceover. That's only because that way you can play around with the volumes. Listen without any voiceover and just hear the sound effects. Do they appear at the right times? If not, you can go back and maybe select a different time or simply move them around like we did before. What you can also do is do it with the sound effects export bring it in again for the voiceover or add another set of sound effects. Because this is an online tool when you add a lot of overlapping layers, it may crash, which is a very normal thing. If it happens to you, don't panic. But one way to mitigate that is to export it and then bring it back in. That way, the first changes you make will be secured and permanent because you exported the video and then you can build it up to as many layers as you'd like. That's about it for the studio tool, pretty straightforward. There's a lot that you can do with it, and when we go back, as we mentioned, we can view our project, go back into it, make additional changes. In the next lesson, we're going to learn how to generate an audiobook using ElevenLabs. This is another straightforward tool. Essentially, you just click upload, choose your voice, and that's it. The next one is going to be podcast, which gets a little bit more in depth, but we're going to look at all of those in the next couple of lessons. 13. Studio Updates: Now, we already looked at how the studio feature works in ElevenLabs, but since then, there have been some upgrades and some new features that can really transform the way you use ElevenLabs. So we're still in studio, and usually it would just be a box where you type in something for voice. But ElevenLabs now allows you to make videos, music, and create auto captions, as you can see so these are just a bunch of examples from what you can do. So a faceless video captions, dubbing, voiceover, video to music, and generating audio. Your projects are all the way down here. If you ever made one, they're going to be listed down below, and you can even filter through them using these tags. When you go up here, you can immediately jump in, say what you want to create. You can think of studio as the all purpose platform. So instead of going through each one, maybe combining two things, you can just go to Studio and get a mix of all of these features. You can attach something as reference from your computer by clicking on this or upload an entire folder using this. Instead of creating a video right here, I'm just going to create a new blank project and we get two options here. First is a video project and the second is Audio Project. The audio project is what we looked at in the previous two lessons. But the video project is that additional thing. So just to go over audio project, I'm going to click on that. And the idea is pretty much the same. So we are in Audio Project, and the main features are pretty much the same. So you get your little workspace in the center. We can break it into chapters. You can think of this as tabs. We have our voice library, some sound effects that we get to incorporate here, music. This is another new feature that ElevenLabs has, and, of course, things that you have uploaded for your project. Could be a reference for a character. It could be an audio sample, anything that you want. They're going to be listed right here for you. Now, you can upload to the project like this. You can directly record into the project, or you can import files. So, you can use all three options. And then, of course, we have the edit tab where all the familiar sliders are going to be here. We get to type things describing basically that final project that we want. So it could be a short story about XYZ, and you just type it in here, and then all those elements come together. It could be a voiceover for YouTube video. You can upload the YouTube video as reference and get the help of ElevenLabs to create an AI voice for you that would match the vibe of that video. We're going to do a very simple exercise right here where I'm going to type in, like, instant of a horror video where our character is a little bit scared. There's going to be some sound effects. We can pull in some music here, get a good suitable voice, and, of course, add breaks and audio tags. Now, over here we have our models. We also have three. So I'm going to use that because of the audio tags, and then we can work around with the type. I'll leave mine for as text, and then you can add in some voice, as well. So let's just explore the library, something that will work well with, like, a scared character. In the theater of life, we are all actors sharing the same Just trust yourself. Then you The thing always happens that you as we are liberated so let's go for default voices and look for scared. Life isn't about finding yourself. The years teach much, which the day? It is not so important to know everything as to appreciate I have never been hurt by anything I didn't say. We make our own fortunes and we call them fat. God has given you one face, and you make yourself. Gratitude is rich as complaint is. So I will go with Eric for this case. We're just going to click on that name. There's going to be a checkmark, and we could just go back to Edit. So now we can see Eric as our voice. We have version three text, and, you know, everything else is good so far. I'm now just going to start typing something very simple. So maybe, like, I think I heard something. It sounds like it sounds like something is coming. We can add in some audio text here. So but heavy breathing. Oh, no, no time. Now, here I'm using the three dots, but if you're using another model like the V two, you're able to use the audio or the speech breaks, actually. But if you want to use V three for something more natural, you could just, you know, do the same thing here. So that's our first line, and here I'm going to put some Erie music. So we have some songs here, but we could also generate our music. So I'm going to explain this further in the later chapters. But for now, just describe the music that you want. So maybe Erie suspense background music. And I'll just leave it with the current settings. We're getting three versions, which is a lot, but you can maybe change it to one, so it focuses more on the first one. Choose your model right here. And then the duration, maybe you can do like 5 seconds or something. If you want a zinger in there, there could be, you know, someone speaking, and then we can fine tune it further. So now, when I clear my search, we can see the three generations that were made. I'm just going to add the first one in. It's importing the music. We just have something for the background, and we get this little timeline. So I get to maybe have it start here, extend it, change my music if needed, and build this audio from scratch. I think I heard something. It sounds like something is coming. Around here, I want him to say something else. So let's, like, go to the next line. We're like, Oh, what is that? And then we'll add a sound effect here, which is another feature. I'm gonna do, like, rustling in the bush. I'm going to do something very short. Maybe like 5 seconds. So I'll just add this here. Importing, we have another element now, and you get to just build upon this. I think I heard something. It sounds like something is coming. Oh, what is that? Maybe a little sooner. Oh, what is that? Adding more text. And then let's say it's like a cat that comes out, so we can do something like Alright. And then while I'm here, I'm just gonna do cat. We can see if there is something. Let's do cat meow. Alright. And there's our final sound effect. Now, I just got to clean this up, and we have our first updated studio project. Now, before I generate this, there are some things you could do with each of these layers. The first thing is to cut something if there's too much of it. For example, right here, I don't need all of this audio, so I could just zoom out right here and pull the end of this whole audio. You could kind of counter this by generating something shorter. But this is also another way that you can cut things. And another faster way is to just grab, you know, your segment. Another thing you could do is just click on that Audio segment and hit S on your keyboard, and it's going to split it. You can simply delete that extra bit with the backspace. Next thing is to work with the volume and the fade effects for each of these segments. So, for example, we had the meow. It just starts out of nowhere. So maybe I could kind of have this audio fade out and the audio here fade out as well. So basically, it's going to lower the volume to zero over a longer period of time. I think I heard something. It sounds like something is coming. Oh, what is that? Oh, no, I can't handle this. Oh, it was just a cat. Silly me. And for some reason, we have a female voice here, so let me just change that back. You can change the audio just like that. The cat meow is a little too loud, so let's just bring that down, and we're good to go. I think I heard something. It sounds like something is coming. Oh, what is that? Oh, no, I can't handle this. Oh, it was just a cat. Silly me. And that's the updated studio interface. Again, it's not that different from what we looked at, but there are some additional things like music and this timeline that we have right here, and you get to import different parts of ElevenLabs into your next audio file. Down here, we have a bunch of other AI tools that you could utilize. So maybe enhancing text. If you want to write something a little better, you can use the enhanced text feature. So I'm just going to highlight that. Click on this. Could do that. If you're importing audio and there's background music, you can use this option right here. You can change the voice with voice changer. So maybe change the pitch, the speed, and all of that. And you can also direct the speech with your voice. So basically, if he's speaking in a weird way, I could record my voice with the way I want him to speak, and then 11 labs will just follow through. Again, you can export very easily with this button. You can share, name your project, and create different chapters if needed. 14. Audiobook Tool: The next tool that we're going to take a look at is how we can make an audio book. When you click on it, you can straightaway upload your short story. I have a short PDF for you guys in a resource pack. It's a children's book by Monkey Pen. They make free children's book available for you. You can upload anything with these formats or even write your own story. So let me just show you what the PDF looks like beforehand, and from there, we can understand how ElevenLabs converts text into audio. So this right here is a short story that I'm going to upload. This is monkey pen. You can see that there's a lot of logos, a little preface text, we get illustrations and all of that stuff. So how does ElevenLabs turn this into an audiobook? Instead of copying all of these, all you have to do is upload it here. Once you've done that, you can remove the document to upload something else and immediately choose your voice. Go over there. You can preview them and look at the different categories. There is no great the years Our distrust. Love. Life without love is like a Just trust yourself. So I'm going to go with Alice for this one, click on it. Then we have this option for auto assigned voices. Essentially, that's going to switch between characters. This story doesn't really have dialogue, so I don't need to do that. But if you wanted to just turn it on and you can see that it's going to take a bit longer. Simply hit Create and you're going to be brought into this page, which is pretty similar to studio. We get all of our text here, and then you choose your voice and add the sound effects as we learned before. These are the preface stuff, and I could simply delete them so that the story the storybook starts right away. Let's go ahead and hit Generate. Command or Control A, so you select everything, your voice, confirm it, then begin the generation. So I generated the audio. First thing you can do is, again, you can change the name, and then we can just hit play and see what it sounds like. Hi, I'm Professor Moisture, and I will be telling you about water. You can call it rain. You can call it snow. You can call it sleet. You can call it hail, but it's water all the same. Did you ever wonder how old water is or where it comes from? The answers may surprise you. The next time you see a pond or even a glass of water, think about how old that water might be. Do you really want to know? I thought you did. Did you brush your teeth this morning? Well, some of the water that you used could have fallen from the sky yesterday. So as you can see, does a pretty good job with the capitalization. I did not do that. It came straight from the PDF. And even the way the voice pronounces this sentence, it's pretty good. And her voice in general, works perfectly with the children's book. But again, just like last lesson, you get to add pauses or even go to the settings here to add more exaggeration to 37%, lower the similarity. We slow down her speech a little bit and at the same time, I'm going to add some pauses in between these sentences. Let's do a pause here and that's about it. Let's do another Commander Control A and hit generate. Let me actually save this, then generate this. Hi, I am Professor Mois Tui, and I will be telling you about water. You can call it rain. You can call it snow. You can call it sleet. You can call it hail, but it's water all the same. Did you ever wonder how old water is or where it comes from? The answers may surprise you. The next time you see a pond or even a glass of water, think about how old that water might be. Do you really want to know? I thought you did. So it's pretty straightforward, works just the same as the basic tools that we learned about already. When you're done, you can hit Export and either publish it to one of the platforms that you see right here or simply export it as the single file or the paragraph separately, the audio format, and there you go. This is where it tells you the amount of credits it will use. So all of your credits are over here and yeah. You again have that timeline down here, zoom in and out, delete parts, add to it, and you can just bring your stories to life using this tool. Let's move on to the next lesson where we see how we can create podcasts with ElevenLabs. 15. Podcasts: Our next tool is creating a podcast with AI. When you click on it, again, it's like the audiobook. Straightaway, you can upload your content, import it from a URL, or even use a existing project. If that's the case, you get to select it like that and then put in your voices. But let's start over here, which is where you get to upload your document. Now, one thing that I'm going to do for the purposes of this lesson is use HGPT to make us a sample podcast. In your case, you would be having your own file. If you've written a script for your podcast, you just save it as one of the formats down here and simply upload it. But I'm just going to go into HGPT and ask for a sample podcast with a host and one guest where they discuss the weather and how fun Summer is. Make this 5 minutes long. So we even got a name, which is pretty cool. Our first the host is named Alex and the guest is Maya. I'm just going to let this generate and we're basically going to copy and paste it into a Word document, and that way we can save it to one of the formats. I'm just going to open one here, a new document, and just copy the podcast itself. Maybe all of this stuff because ideally in your sample, you do have a bunch of information as well. There is my podcast. Let's save it. Name it my podcast, save it as a dotxem. It's save and upload. There it is. You can remove it to put in another one. The next thing that you can do is choose the format of your podcast. Focus podcast with key updates, casual conversation between a host and a guest. We did ask for two people in the podcast. This is what I'm going to go for. Then we get to choose the duration. I'm just going to go with default, then hit X when I'm done. These are the voices for our people, the host. A single rose can be my garden. And then the gift. If you spend your whole life waiting for the storm, If you go on any of these, so let me just search for Jessica real quick, you can see that one of her tags is conversational, and I did ask for a podcast where two people are having a conversation. You can also look for other conversational people and then make your choice from there. Hi there and welcome. Nicht to the Dao. The default ones are by ElevenLabs. They're pretty good usually. Nature as a mutable cloud. We make our own fortune. But I'm just going to keep the ones that ElevenLabs chose for me. Down here, you do have some settings. So what is the podcast focus? I think we said it was hot weather. Let's add another highlight summer. It's safe. So what it's going to do is that it will emphasize those words within your podcast, and I'll just show you what that looks like. Lastly, we have the language. I'm going to stick with English, but we will come back to try a different one. Let's generate and see what our podcast looks like. Okay, so there is my audio. And we have 6 minutes and 26 seconds. That's because it has the pauses in. But if you saw that it's way too long or way too short, you can always go back to Cha hBT and ask for it to either make it 3 minutes, 20 minutes, however many long you want it. Let's go ahead and listen. Did you know that heat waves now kill more Americans each year than hurricanes, floods and tornadoes combined? The way we experience summer is fundamentally changing, and today we're diving into how communities are adapting to these rising temperatures. Those statistics are truly eye opening. I've been reading about how different cities are developing innovative solutions to deal with extreme heat. Well, what's fascinating is how this has become a major public as you can see, it does switch between the people, and down here, we're actually seeing the colors assigned to Chris and Jessica. One thing that you may have noticed is that by default, ElevenLabs removed the first part of our script where it just talks about the description about the podcast. So it cuts immediately to what these people are saying. I'm not seeing guess parentheses Maya or the podcast title or anything like that. That's pretty impressive. And what you can do is alter the voices at the same time and have it regenerate everything. For Chris, I think that he could talk a little slower and be a little bit more stable, actually a little less stable. He sounded a little bit too robotic and a little exaggeration. Let's do something similar with Jessica. I think she could definitely slow down with her voice. I think we didn't save. Chris actually lower this and make this a little less stable. It's safe. One more thing that I do want to do is add in a few words of, like, pauses, human errors to make this poadcast a little bit more realistic and that's something ChatBT could help you with. I'm just going to paste my text. Take this text and add and pauses to make it sound. More human. Also, let's reduce this to 3 minutes. Quotation marked, paste and quotes, and then hits Enter. Just a disclaimer, this is by AI, the information here may not be true. So just take all of this with a grain of salt. If you want to make statistically correct podcast, be sure to choose reliable sources, but this is just for demonstration purposes. These numbers, these places, programs, they're probably not real. Remove the titles for host and guest. I just want the dialogue separated by paragraph breaks. All right so this will work just fine. You could have it do that to remove the titles for host and guest, but then you would have to assign Chris and Jessica to each of these sentences. What I'm going to do is just copy the previous response and then re upload it into our project. Command or Control S, let's go ahead and first of all, delete everything and choose a document. Let me actually go back here, start a new one so I don't have to edit anything. The same settings as before. I will just add the highlights here, maybe summer. Here, what you want to do is actually look at your scripts and find something that you'd like to emphasize. I see kids here maybe, maybe temperature. If it turns green, that works for this focus because it's in the script. I think I'll just go with one since there's a lot of different information being passed around here, but temperature is definitely something we're talking about. Save the changes. Let's put this on English. The voices are fine, and start generating. Okay, let's play and see what we got this time. Here's something that stopped me in my tracks. Heat waves are now killing more Americans than hurricanes, floods and tornadoes combined, and it's completely changing how we live in our cities. That's such a striking statistic. And what really gets me is how this isn't some distant threat. It's affecting communities right now. Okay, so I have my awards right here. It kind of brought in the last script, but I could just switch it out like I just did. Hey, um, did you know that heat waves now kill more Americans each year than hurricanes floods and tornadoes combined? Yeah, it's wild. The way we experience Summers, So if that happens to you, just go back to your script and copy what you want. And here, I'll add it to what was there before, paste it. And what you need to do is highlight this entire section and only apply Jessica or whichever voice you want. And this way you can switch it out. But now that there's ms and pauses, it sounds a lot more human. That's exactly why we went to ChachipiT. Once again, I will lower down his speed a little bit, add more exaggeration, save it. For Jessica, I think she's fine. Maybe make it a little less stable. Her speed is perfect for a podcast. But once we're done, we can regenerate just a bit of it so we don't use a lot of credits generate. Hey, did you know that heat waves now kill more Americans each year than hurricanes floods and tornadoes combined? Yeah, it's wild. That stat is honestly kind of scary. I've been reading about how some cities are coming up with these really creative ways to deal with extreme heat. The way we experience summer is, like, fundamentally changing. And today we're just diving into how communities are actually adapting to these rising temperatures. That's such a striking statistic. And what really gets me is how this isn't some distant threat. It's affecting communities right now. Alright, so there's our simple podcast example. Let's go back to the first menu and take a look at what these are. So input from URL could be like a blog post, something on the website, on your social media or something, and it's going to use this technology to make a podcast that you can listen to. I will try this with a blog post maybe and then see how it uses that technology for it. I will find a URL and then I'll be right back. Let's go to Wikipedia and get a very simple article. I will search for maybe apples and simply paste this URL into the bar here. Then again, you can choose which type of podcast. This time I will go with a bulletin because it's facts about an apple. Let's make it really short. A single rose can be my choose a different voice this time. God has given you one fame. Just a British voice. Language. We can do English. And then we can see what that sounds like. Okay, here is our Apple podcast. Let's hear it. Apple's remarkable journey began with its wild ancestor malls ciersi in Central Asia, evolving into today's malls domestica through millennia of cultivation. Commercial orchards rely on clonal grafting rather than seed propagation, as seeds can produce unpredictable results. This careful cultivation has led to over 7,500 distinct cultivars, each developed for specific uses from. So as you can see, I'm not going to play the whole thing. It basically summarizes entire Wikipedia page and gave it a more conversational speech. So the way this started is that it talked about, you know, the remarkable journey, which is from, I believe, the cultivation section of the the history section, actually. Started from here because it knew that you're going to need some context. Then it went into the ancestry, when it was sequenced, even the genome, which is a whole different section. It's actually pretty impressive how it takes little bits of information from an entire web page, such as Wikipedia, pretty large, and it makes it into a podcast. That's pretty good. There's always you can change the voice, add break, sound effects, or anything else that you want. But the last thing that we want to try with the podcast is existing project. Right now I have four in the background. All you have to do is choose one. Choose the format. I will make it into a conversation, short, choose my host and guest. She tore her gaze away from her ruins to Arabella and Adam. People who are crazy. Let's make it Japanese and see how the different language works. So as you recall, my scary story was just me writing a bunch of sentences. We added some sound effects, but there wasn't much. Let me just show you real quick what we had. Not a lot of sentences going on. I'm going to have a podcast be made out of this. So let's choose our project conversation, short, and I'll choose my voices again in Japanese. So here it is our story in Japanese. I'll give you a preview, and then we'll turn it back to English so we can see what sort of changes ElevenLabs did. You show sakaka dee tomo Koa Nova Mirena, kommen Kiowa, Tata tore Umacho So done. Hi, ShotunaaKo jo too. So great conversation right there, I'm assuming. Let's go back over here and do the same thing, but in English. That way, we'll know how it witches. I'm going to use the Japanese version as my project. Same settings. Well let's actually go back to our chosen ones, make it English this time, and then we can compare. Here is my first of all, converted text. You can see from our short story that I wrote, it turned it into a conversation. Let's go ahead and hit Play and see what we got. Ever wonder why the scariest moments in horror aren't the monsters, but those quiet seconds when you're completely alone. There's something deeply unsettling about true isolation. That's exactly what makes this story so effective, the abandoned factory setting where literally no one is breathing. So we took our very simple story. I'm gonna open it for a second to show you so this couple of sentences, I'm just describing this video, it's basically it made a podcast based up of this story. So these two people are talking about our character lost in the Forest, but you can see what a great conversation was brought up all using AI, and we can switch between the different sentences. Podcast feature is really powerful and there's a lot that you can do with it. In the further chapters, we're going to do a podcast from scratch. Don't worry if you didn't get a lot of practice with this right now. We're just going over the basic capabilities of each of these tools so that we are ready for those interactive projects. Now, that concludes the studio section. Import from URL just lets you put in a Wikipedia page and for example, it could be a blog post, Wikipedia, anything, and it's just going to simplify it for you. Whether it's a podcast, a audiobook, a narration, it doesn't matter. ElevenLabs will decide what works best and will give you an audio instead. This is pretty straightforward. You just paste your URL, choose your voice, and that's it. If there were a conversation involved, you can just turn this on. But it's a combination of the tools that we already looked at. I will not go over this right now. That is all for studio. In the next lesson, we're going to go over dubbing, which is basically where you get to put a audio on top of an existing audio to, as it says, localize the content. Currently, there's 29 languages, and I'm pretty sure in the following months, there's going to be more. Let's move on to dubbing. 16. Dubbing: With dubbing, you can translate your original clip using one of the 29 languages that are available and basically decide where the voiceovers go. So this is the interface right now. There are some tutorials for you to look at to get started. But the first thing you have to do is simply click on Create a new dub. When you do that, you're going to be making a dub project. So I will call this my scary movie dub. The source language is the language that your original video has, for me, that is in English, and then you choose your target language. I think I will try something, let's see. Let's do Spanish, for example, click away, and then you can put in your audio or video source. Apart from your classic upload button, we have YouTube links, Tik Tok Links, other URL, and manual. So if you want to do manual, just to walk you through this, this is the video file. A CSV file is going to be your transcription file. Foreground audio file is going to be the speech, whatever that's in there, and then background could be a little music or something. So you would only come here if you have the separate files for each of these. If you don't, it's the same thing as doing all of that in one file here. So I will upload this video, which I'll show you guys in a bit. They say she woke up with blood in her eyes, wrapped in metal and silence. The plane was gone. Everyone else gone. So this is a video that I made with Runway AI, which is another tool, as I mentioned in the previous lesson. But this time it comes with audio. And as you heard, it's in English. So I want to turn this English audio and have a Spanish dub on top. And this video is available for you guys to download in the resource pack. So let's go ahead and upload it right here. You can preview it once again. They say she. Then when you scroll down, you can delete it to put it in another one or simply continue. If you create the dubbing project, you're going to be able to get a timeline where you adjust the different audios. Definitely turn this on unless you have a very straightforward project, and it's just like a plug and play thing where you don't need to edit any. Number of speakers, it can detect it for you, but you can make sure it does it correctly by choosing one of the numbers. I only have one speaker, so I'm choosing one, and then the time range to dub for me, it's going to be the full thing. 02 53 seconds. Disabled voice cloning, turn that on so it doesn't make an identical voice. That's it. You just hit Create Dup, it's going to upload it, and now it's going to convert it to Spanish. It's using the playground tools and some of the more audio into Spanish, maintaining the natural tone, the mood, and all that stuff. I will let this run and then I'll be back so we can see what we have. As you can see, it's done, I can simply click on it and move into my project. As you recall, we enabled the dubbing project and that's why we have this option. If you do not enable that, you will not see this interface. So here is the transcribed audio. You can see the text, and it does match to what we hear in the video itself. So I don't have a lot of information, a lot of audio in this video. It was like 53 seconds, I believe. But right away, we can see that it separated the background music, the foreground music, and then the speaker. So right now, I'm not going to do anything and just simply play it for you guys. They say she woke up with blood in her eyes wrapped in metal. So you can see it's still the original video, and what you can do here is play around with the different audios it extracted, add something, remove it, and do a bunch of stuff. But before I get into that, let's first look at this interface and see what's happening. On the top left, we have the transcription of the audio, as you can see, can also click on this so that it redoes it for you. That's about it for this section right here. On the right is the preview of the actual video. If I move my playhead, you can see that I'm going back and forth, and that is synced with the transcription. Speaking of the timeline, we have layers. These are three layers right now, but if there's nothing on there, that means that there's no foreground music right here. This is the background music alone, and this is the speaker speaking. You can see here the speaker doesn't say anything, but the music is still playing. Let's hear what that sounds. Not empty. Every steps. You see it says something, pauses, says it again, but there was still music playing in the back. Over here is the names of the layers, and you can change the name, not for background and foreground, but for your speakers. If you have multiple speakers, you can give them a name and this will keep things a lot more organized. I can call this Jake. It's labeled original because we're still in our original video. Everything you do is going to be saved automatically, so don't worry about there not being any safe changes. On the right, you can only enable that layer. If I click on background, you can see that it grade out the foreground and Jake. When I hit Play, even though I'm on this part, I'm not getting any speech. Let's do the same thing with the Jake audio. And the trees. She walked for hours. It's just our speaker. You can see how clean it separated the two. I only gave it one video. On the side, you are able to delete the speaker and add a new one. You can use this to go across your timeline. On the top right, we have how many seconds you've elapsed. We have the select, which lets you select things. We have the trim if you wanted to trim something. I'm going to select the background, hit the scissor icon, and then you can see that I now have two sections of the same audio. Go to hit Command or Control Z to undo that. That is only one on this end, we have add clip so you can add up different video if you want. I'm not going to do that, but if you did want to, you can decide where it goes, maybe here and then maybe turn this into an Saffx or something. We'll get into that later. To delete, just select it and hit Backspace on your keyboard. Now, with every one of the audios, you have the option to change the volume. This is what it sounds like right now. Let's only highlight the background music. Now let's go back and reduce the volume. We can barely hear it. If you see something is too loud, you have the option to adjust. All right. That is the basic interface. Over here, you're able to change the name of your dubbing project. Down below, you're able to see how many credits you have, and this is where all the dubbing happens. You have the original, which is the original clip. If you go to the Spanish version, we're going to get Jake in Spanish. Here's how it's translating the original audio into the Spanish version. Let's see what that sounds like. Inches Esperto consangre L o Sojos and well t. Let's actually increase the audio for the background music. DenchesEsperto, consangren Los Sojos and Welt metal is IlensioEabon, Able Zaparsdo, Todos Los mas, desaparzidos, solo Aga, Ilosarbos, caminodor teoras, Talves dilas, lb so I don't speak Spanish, but if you do, you can see how it translated these word by word. If for some reason you saw that it's missing something or it didn't translate right, you have the option to change things up here. I could just delete that word, for example, and quote something else. When I do that, I can hit Generate audio again. If you go to Jake Spanish, there's this gear icon, and this is where you get to choose the voice and the familiar settings that we've been working with so far. First, if you go to voice, you can choose someone here. Let's see. Not a clone. I have crossed hope. Let's actually look for Spanish. Be ambitious in your goals and diligent in amateur voice. Let's, um, try to do, like, a male. Let my voice be all right, this is pretty good. Let's add it to our voices so we can access it easily. Choose our model. I'm going to go with the recommended and then if we want, we can change these sliders. I'm just going to keep it at default and then close this. That's the control you have over the voice, and then you are able to add even more languages. If you want to do a Spanish dub, a Japanese dub, you are able to do that easily. Let's do a Japanese dub. It's going to process everything. Immediately, we can see the Japanese version. That is looking pretty good. Metals lensoabon Able So that's in Spanish. Let's sit still and get the, you know, generated audio for Japanese. Gabe, KinsoctoKs in its marite Mesa metato, Yareto. He co So I use that new voice. I believe we chose Christopher. That's why he's sounding a bit different. Now, you are able to create a voice from the selection or generate audio. These are the familiar settings. Going to see if I can extend this a little bit. That's a volume from your original clip. If you grab something like this, the language that you're currently on, you can create a voice from this selection only. So if I do that, let's give it like Paul. As a name, create. Now, whatever adjustments I did to this particular clip becomes a voice model that I could use with other tools within ElevenLabs. Just like how we have these voices, I just made one called Paul, going to see if I can access it from here. There we go. There's my Paul that I just made and you can create as many as you want. Now, for voice settings, you can inherit the track settings, which is whatever you do in the track is going to be applied. If you disable this, you can do what we were doing in this skeer icon but over here. Choosing your voice, your model, and then configuring the voice with these sliders. Now, you're also able to dictate it yourself and it will follow the way you're reading it to adjust that voice. But you can also have your own voice playing over this if you want, and that's all done with this button. This clip is only 20 seconds. That's why we see it right here. Whereas the clip the entire movie, let's call it, is a little longer. Okay. Then down here, we have a bunch of other stuff. First one is add dubbed speaker Track. I could add a different speaker. You can see we have another one. Add voiceover track, which you can do with the text-to-speech, for example, like that, let's increase it. I don't have any audio right now, the source, so I could put something, something like that, let's generate. Now because I'm in Japanese is in Japanese. Oops. Let's only highlight this. H, I'm lost. This is the original, but we can do the Japanese version. I stretched it out, but it's really short. Going to hit Generate audio again. Myota. There is our short, Oh, I'm lost in Japanese. Can have both of them play at the same time. If you wanted to, let's move them around. Ho, I'm lost. Myota. I could just switch between languages and go from there. You can extend like this to make it a little slower. It sounds like a monster right now, but the more tight these bars are, the more natural it's going to sound. L. So I make it really tiny, it's going to be super fast. Lyota. Then we have add SFX Track, which is your sound effects, which we already know how to use. When you hit Plus, you're going to get a new layer, and all you really have to do is first of all it's enable sound effect, go down. Let's first actually make a space for it and describe our sound. Let's do a parrot. Generate the audio. It's showing the SFX in English and Japanese, but it's the same thing. That's our parrot. We can alter it. We don't get different samples like we usually do with the sound effects panel, but if you don't like this, you can hit regenerate right on top or delete it with backspace and start over. You can also do a longer one, a longer sound effect. Let's do raining or just heavy rain maybe. Generate audio. The reason why you see both English and Japanese is because that's going to be copied into all of these guys. I don't know how that sounds like heavy rain, but you get the point. You just have to be very specific. Let's try to do heavy rain pouring down on Stone Road with thunder in the background. Something more specific. And there we go. Sometimes, you got to make the prompt a little longer. Then I could adjust the volume like we did with the other ones. And when I have everything play, this becomes really important. So let me just go over, you know, this section where we're getting a bunch of things. Mechintakega, NokoTta Kano JawananjiKamo, Aukiso, Ariwa Nanihim. So you see how this needs to be a lot lower because we have our speaker negative 27. Amo, Arakio. We actually have to alter the Japanese one. Amo Aukisok Ariwa NanihimKamo Shini now we can see that it's a lot more natural. You can go to the Spanish version or stick to the original. The last thing you can do is upload audio. You can upload your very own sound effects, your own voice overs or anything else that you want. If you made an audio with a different tool, you can export that and import it right over here. Existing music, sound effects, backgrounds, and all of that stuff. Now, keep in mind that the speakers won't be detected. The only way to bring in speakers is through the original way that we did, and that way, ElevenLabs is using the correct technology to do this whole dubbing process. Okay, once you're done, let's say this is my finalized project, you're able to hit Export down here, export it in whichever language you want, the ones that you made down here as the output that you want. You can do the video and the audio, just the audio, only the caption, transcribe, a zip file, anything that works best for you. So let's hit X, and that's about it with the dubbing tool. It may look complicated when you first come in here, but it's pretty intuitive when you look at it. You have different sounds on different layers, different languages well organized down here. Got a preview. You can alter the text, and it's a pretty handy tool. Now that we learned about this new tool, we can move on to the next one and see what other features ElevenLabs has to offer. 17. Speech to Text: Speech-to-text is another great tool that you can utilize when it comes to ElevenLabs. Now, it's very accurate and it's getting better as we go on, but this could be used for meetings that you want to transcribe or a separate audio file that you want to just convert into a different format. Now, the way it works is that you have different transcriptions right over here and different speakers. Let's say you are always using ElevenLabs to transcribe meetings at your company. There's always a certain number of people, and you know everyone by name. You can add those speakers, those participating members as speakers here, and ElevenLabs will mark that for you. So when you put in a recorded meeting file, it's going to identify the speakers and then mark that for you in the transcription file. So that's a really cool thing that it does, and you can get started by clicking on transcribe file. This could be different things. It could be something you upload from your computer. You could record something yourself right now. You can put in a YouTube URL or any other URL. You can have the language said right here, so many options, scroll down and choose the language of your choice. And you can also have 11 labs write down the audio events. And the way it's describing that is that if someone's laughing in that audio, it's going to put laughing. If someone's crying, it's going to put crying, and that's going to be included in the final file. If you wanted to ignore those things, you can just turn this off. But it's a pretty cool feature that it has. Now, the file that you're going to get can also include a subtitle. Let's say that you want to transcribe an audio and then use it for your video. Instead of putting that back in for a different subtitle file, you can have it all be done in one go. That's a separate thing that you can turn on. And this feature is going to remove everything that's a filler sound. So if someone's coughing, if there's a pause, any ums, you know, these voices are going to be removed, and 11 labs can ignore those and not include them in the final file. Now, we also have the assigned speakers from library. These are these speakers that are going to show up from the tab that we looked at earlier. So if I turn this on, I currently have no one saved, but we're going to see what that looks like once we add some speakers. We have some key terms. Finally, we have key terms. Say you are working on a very important project and you want 11 labs to make sure that it doesn't miss anything with the word project or something specific. You can add those key terms right here, and 11 labs will take extra caution to bold those items and not miss them. So if you don't put these here, it may miss a couple of things when they talk about that boosted term. But that's only if you have a boosted term. If you don't don't need to put anything in here. Let's get out of this and take a look at what the speaker looks like. We're going to go there at speaker, give them a name, speaker ID, and an audio sample. So this sample is how ElevenLabs is going to detect who is speaking and will tag it with their name in the transcripted file. What we're going to do is first get an audio. For this, I'm going to use this video. This way we can all grab the same clip, but feel free to put in your own audio files as well. Because we are able to put in YouTube links, we can easily just grab the URL right here, copy that, and paste it right here. So just switch to YouTube. Past it, and we're just going to change a few things. So this is in English, but if you leave it detect, it will do that, you know, on its own. I could turn this on or off as we said. I'm just going to turn it off. Say, I just want to get a quick summary of what happened during this meeting, and I just want to remove anything that's unnecessary. Now, we can add the key terms right here. I'm just going to take a look. I think we're going to talk about meetings here. I've heard some of my mentes talking about how it's really hard to get out of bed on Fridays. So we're just going to focus on the key term bid. Say, that's, like, something really important for me. I'm just going to put this right here. And if you hit Inter, you can add more keyterms. I'm going to skip all the way to the end. Let's get another key term. Are you meet with him. It's a lot to deal with for middle schooler so middle schooler. I'm going to put that in as well. Alright, so we're going to start transcribing this. It's only a minute long without any assigned speakers. And then I'm going to show you how to do the speaker end as well. So let's click on this and have it work on this file. Now, if you want and you have certain people in mind that you want to save, you can use the speaker feature right here. Add those people in, and those will be kind of bolded throughout the transcribed file. And that way, you know, when you have a meeting with 20 people, you can easily filter through and find out what John Smith said during that meeting. Only thing that you are going to need is a sample of their voice alone. Now, because this is a YouTube URL, I can't really, you know, put the whole thing. But if you have that person in mind, you can just put in their audio alone. So that's that. Let's take a look at what the transcribed file looks like. So there we go. We have our speakers listed as speaker, but we can now switch their names out and then later save them to the library. So you can either start with the speakers saved or save it from a video like this. The first one we found out their name is Mara. I just scrolled down on this video and someone has it all transcribed. So I'm just going to copy that over and, you know, change the default things right here. So we have our first speaker, and I think this is the same one, so I'm just going to, you know, click on it, and immediately it did update it for me. The first speaker is the next speaker, I'm assuming is this person. Oh, we're actually getting some cues, so glasses is f. I'm just gonna switch out these names so we can save them to the library. So speaker two, who's this guy? Just skip a little bit here to find out. So that's her. I'm assuming it's this person. We're just gonna put that in as Speaker two. Alright, click away Speaker three. So this person must be Thomas, and then we have Matthew. And we can see that 11 labs did detect five people. It's pretty accurate right now, and I have these speakers now saved. Now, I could add a speaker, say it forgot to add someone, but usually when the audio is clear, it doesn't do that, but just in case you can come in. And if for some reason the speakers were mixed up, you can grab the edges right here and decide or maybe fix the duration of each person's speech. So that's that. I'm just going to undo what I did by clicking on that button. Now, on each of these guys, we can change their order, but these orders are based on the normal flow of the video. So unless it's wrong, you don't really need to change them. We can also change the color of the icon. If you want, just to set them apart. And here is where we get to add them to the speaker library. So I could give them a name. And now that it has the sample from the video, I just need to give it an ID and a name. So I'm just going to copy this person's name and then put it right here. So the next time I have another meeting file with this person speaking, it's going to recognize that it is Maha, and it will hold that in a interface like. Alright, so we're going to check all the text. There's also a spell check feature here if you don't want to, you know, read through everything. But let's take a look at what's on this side. So first of all, is the name of this file, this audio file, which is just the UT BRL name. I'm going to leave it as that. Language is English. I did not add any subtitles, but if I wanted to, I could just, you know, that and I'm going to get a separate subtitle file. So that's something you can explore if you need to. We have a spell checker. I click on it, it's going to go through every word in English and let you know if there's any errors. Thankfully, we don't have any of those, so I could just collapse this. Next, we have the segment properties. So this right here is the segment, the duration, the start time and time, and that's about. So if everything looks good, all you have to do is go over to either Export it or copy to clipboard if you're just pasting it in a Google Doc. But clicking on Export is usually the way to go. I'm going to click on PDF just so that it's, you know, accurate and easy to share. Over here, you can add timestamps. So at this second, this person said this, and we can now also include the speaker names because we did change each one. Click Download, and now we have ourselves a transcribed file. There we go. First speaker, second speaker, third one, and it's telling me the duration, the end of that segment, the start of that segment, and, of course, what they are saying. If there is a pause, it's going to break it down into multiple segments. So technically, the first few seconds of the video, only Mo is speaking. But because she paused during her speech, there is two segments now. Back here, you can just share it directly on ElevenLabs and not anload a file. So you can search for people on your ElevenLabs team and, you know, create a workspace. Say you're working with different editors and you want to, you know, make sure everything is good to go. This is a cool feature for that. Going back to the page, I should have one speaker saved in my library, and there we go. So I can have the speaker for later use, and if I wanted to delete it, I could just click on this here. Now, apart from the classic speech-to-text feature, we also have a real time model that is still on the demo mode, but we can still give it a go. Basically, you can have 11 labs detect the language and just speak to it in real time. So this could be in between a bunch of people that are speaking. It's going to be all real time transcription. So when I click on transcribe, let's give it some permission, and we're going to have 11 laps take a look at what I'm saying and write it down right here in front you can see that this is pretty accurate. And just to check out the whole multilingual claim, we can switch between languages. I could now switch to, let's say, Turkish and say, Mava, it's going to write it like that as well. But once you're done transcribing, you can click on this button and I could copy this over to wherever I want. Maybe I don't want to spend a lot of time typing. I'll just say it into my mic, copy it, put it into Slack or any other platform. That's the scribe feature. So go ahead and give this a try. It's a really fun feature that ElevenLabs has. This one is also really cool, especially if you can switch to different languages. Although it doesn't currently support every language. If the one that you speak is within this list, feel free to try it and see how it works. 18. Production Tools: Now, let's go over productions with ElevenLabs. Right away, on the left side, you can see that it's labeled with Alpha, which means you're going to need to upgrade or get that specific plan to be able to use productions. So like I said before, I'm not going to go too deep into how to use this because these are human edited tools. And in this course, we're trying to focus on the AI aspect of ElevenLabs. So right away, you can see the services that they provide. We have transcripts, captions, subtitles, and dubbing. These guys are coming soon, but basically what we do and did with AI is now being done by professionals. So think about if you were to go to Fiber to get a transcripted file, you can now do that within 11 labs. As we said before, if you see that the AI tools are just not working well for you and this is a very high quality project, you can switch over to the production services to get something a lot more professional. As you can see, you do get charged additionally, $2 per minute and the prices change. But let's just go ahead in the transcript and I will show you around a little bit. When you click on it, it's the same interface as the AI one. You upload your file. It gets reviewed by the native speakers. Choose your language right here, give it a name, and then you can even add a custom style. Any instructions, for example, really emphasize the word temperature within this transcription or add emotions in there, something like that. We can also add the sounds which aren't words such as coughing or dog barking. If you want to get rid of that instruction, you just click on remove. It's pretty straightforward. The rest of the three things up here are not available right now at the time that I'm making this course. But what they have is that you can tell them what services you want for captions and they will notify you when it's ready. You just fill this out and then talk about your use case. Now, down here, you're able to make a new production and create folders to keep things organized if you plan on using this tool very often. When you click on New Production, it only lets you do the human transcript because that's the only thing that's available. But later on, hopefully soon you will be able to do the other stuff. Captions, a human will be writing your captions, subtitles, another human will be making you subtitles for different languages. They would be the ideal speakers within that target language. Then dubbing, they will translate your work and then record the dubbed version for you. That's the production tabs right over here. If you go to Audio Tools, there are some more things that you can do. For example, Audio native. This one is again with a different subscription. If you, for example, have a blog or a news website where you publish a lot of texts, you would like to make your website more accessible by having an audio version of that text. This is where you would come and if you click on Play Video, they will show you an example. For example, this is a news story, and this is how the audio native would work. Not only are you letting people read your work in a podcast way, but for people with disabilities, they have the option to come on your website and be one of your frequent users. It's pretty cool, and this is what the bar would look like if you were to implement it. That's Audio native. We also have Voiceover Studio, which is going to basically combine a bunch of the tool. We already looked at how to do sound effects and make voiceovers with dubbing. This studio basically combines the two. If you scroll down, we have a demo right here. So we have a voice track and then we can put in our text. It's just like the dubbing studio but for voiceovers. The next thing is AI speech classifier. Now, this one is really cool because you get to see whether or not this audio that you're about to upload was made using ElevenLabs. So if you are suspicious and you just want to make sure that the person who sent you that audio is actually a real person, you can let ElevenLabs find that out for you. You just upload it here. There's not much you can do. The next thing is the soundboard, which we already looked at previously, but I'll go in it once more, takes you to this whole different URL, and it's basically what we did where we combine different sounds over here, we made folders and presets, looked at some of the available default values. Soundboard is free. You don't need to upgrade your plan. Going to the subscriptions real quick, you can see the differences. This is probably the one where you get to use a lot of these production tools. If you were a bigger company, you would go up to scale. This is what I'm currently on, which is the most popular package. But if you felt like you want to use the other ones, simply click on Upgrade and it's going to charge you more to the card that's attached to your account. Okay. So those were some of the production tools that I just wanted to show you that they are available and you are able to upgrade your plan to use them whenever you want it. This concludes our second chapter, which is looking at different ElevenLabs tools. In the next chapter, we're actually going to start building things with ElevenLabs, starting from basic projects and then moving on to more advanced projects. This is a great time for you guys to follow along and build with me and that way you know exactly how to use this tool for specific cases, troubleshoot using the knowledge that you've earned in this chapter and after this course, hopefully you're able to create your own audios for your videos, websites, et cetera. Now let's move on to the next chapter. 19. ElevenLabs Music: A recent feature that ElevenLabs now provides is the music feature. You can now describe the song that you want, and it will generate it for you using different references, and you have so much control than just getting a random music. So this is where you can access it, and immediately, we have this chatbot that we could you describe the song. You can see the templates just coming up. But it's very similar to any other AI generation platform in the sense that we could get different versions, choose our model, set the duration, and then of course, we are using some credits. Down here, if you scroll down, you can see some samples of the music that was made in ElevenLabs. We can take a look at their beats per minute, download them if you want, add them to your favorite list, and even adapt it to your project. So when it says adapt to your project, basically, you're kind of remixing that music for your either voiceovers, videos, images, and all of that within ElevenLabs. Let's hear a few of these samples. So we have a few instrumental examples, but you could also have someone sing in these songs. So that's something I will show you as well. But if you want to quickly browse through these songs and find something for your podcast, you can click on this, and it will list the ones that are most suitable. So what I could do with this one, let's say my podcast is 30 minutes, not 2 minutes, is to adapt it to my project. I would have to use credits for this because someone else make this song, and then I could extend it to 30 minutes. But I could also make my own songs, so I don't have to use, a lot more credits, basically. Can also browse for similar songs. It's going to list them for you based on the texts that the song has and the duration. We get a bunch of different things to look at, and this is all within the marketplace. Now, if you don't want to download anything, you can go to Generations and just simply describe the song. Now you may find it a little intimidating because, you know, you have to put in the right keywords. But the good thing about this is that it gives you a little helper down here where it will generate the prompt for you. So you can either start with a prompt where you describe the song. You can also put in that lyrics that I told you about, add some reference if you want. And if you're not sure how to even write what you need, you can use this section down here. We made these in the previous lessons with sound effects, but they still use the music model. I could further edit them if I wanted to, but first let's do something here. So I'm going to just put it in here. I want a song that is suitable for a log where I am exploring the forest. I want someone to sing in the background. So let's send this and there we go. I now have a prompt based on what I you know, described. So it's a song for Travel Flogs which is true, features finger picked acoustic guitar. We got some birds in the back. And here's the lyrics. So we get some, you know, openings, and then these are the verses some chorus and an outro. If I'm happy with this, I could let it go and press on the generate button, but I could also come in here and make changes. You can either type directly, add something in here, or have it regenerate the lyrics for you. And as I mentioned, you could add a reference here. So this could be another song that you really liked, but I'm just going to let it be like this. And down here, we get to choose the duration of the song. Because I have some lyrics over here, I can't really do that because it's predetermined. So this is a few seconds for each part. I'm just going to set it to Auto, which is however long 11 laps determines. Down here, you can choose the number of variants. When you add more variants, it's going to take more credits. I'm just going to do one for now and you can see that it just becomes 900/minute. We have some fine tuning as well that you could add, so you can go for a different style of singing if you want. We have male, female. You can preview them as well. Have male or female, and you can add those as well. I'm going to leave mine as it is, nothing added because this is a very low stake song, and I'm just going to click on Generate. So once you generate it, it comes to the side, and we can see them being built. As we speak, it's a minute and 48 seconds. Let's hear what it sounds like. Steps on the mars. Softly, softly, green canopy high above M I rivers. So that's our song. If it's, you know, too upbeat, I could go back and maybe make some adjustments. But overall, you can see that it does fit the travel lock style very well. Now, if you like the song and you want to make further adjustments, you can just go to Edit Song, and it's going to load the project for you. So the name of my song is lost in the wind. You can see the history of the chat. On the right side, we have this little timeline where I get to play around with the verses. We have a play button. We can work with the different lyrics and the different tags, add more stuff if we want, maybe we want to add some maybe two more minutes of this song. So you just put it in this chat right here, make it longer, and maybe add a second singer in there, whatever you want to do. Since there is a chatbot here, you don't need to do a perfect prompt. As we saw, initially, ElevenLabs just took my idea and turned it into a decent prompt. Once you have your song and you're happy with it, you can download it on the right side, or you could publish it on the platform itself. So this is where people can use your song and then you earn something. I think the rates differ depending on how many you've made, how many songs you've made, and how many people download your song. But it's a pretty cool option. You can just play around with these, make some decent music, and have it available for other people to download. We can also share this, share it as a video, give feedback. And these are, like, different segments of the song, so we could cut it short or make it longer, whatever works. Since I just made an adjustment, we would have to basically update this song. So it's pretty straightforward. I'm just going to do a slight change to the song. Say, I don't want it to be, you know, this chill, I want to do something with fine tuning, maybe like a rock kind of style. I could do that right here. So let me just undo this extension, and we're going to add a fine tune. Let's go for the country. Since rock wouldn't really fit this style, and I'm going to say introduce a female singer after the course. So that's our version one. We're now going with a version two, and I can see that there is some suggestions here, so I'm just going to hit tab, and that will be added, as well. So let's put this in again, one variation so it doesn't take too long, and I'm going to send that in for 11 laps to update my song. What it needs to do is that it needs to add more lyrics. So here's my female vocal solo, right after the chorus, as I asked, and here is the section. So let's hear what it sounds like. I'm going to skip to this part. So right now, we just added these things, but I'll just, like, copy this part just to show you that you can change the lyrics yourself. So back to the start, let's make that adjustment. And let's see what we have. To the s back to the star. And there we go. We were able to make a song out of one prompt or should I say an idea? We added another singer. We added our own lyrics, and we had this country style music at the end. So go ahead and try this, make a bunch of songs, and it would be a good idea to publish them on the marketplace. You never know when someone might just download your song and you'll earn a few credits on the way too. So if you do want to explore this, just click on Get Started, get yourself a nice profile, and then your songs will go onto the marketplace. This is a great feature since ElevenLabs also introduced video and images. So you can now do your voice overs, background music, the content, the visual content, captions, and basically anything else you need with content generation. 20. Voice Design: The voice library offers so many different voices, all that are very diverse. There's different languages, different tones, styles. But sometimes it's just not the one that you're looking for. But, luckily, ElevenLabs allows you to create your own voice using voice design. So right now we are in D voices tab, but we can already see all the different voices in the different languages. So right now, I can easily go in, choose a tag that is related to what I need and download that song, use it, change the speed, tone, and, you know, go about making my content. We're going to explore this feature right here, which is when you create your own custom voice. And the cool thing about it is that once you have your own voice, you can actually upload it to the marketplace where others can use your voice and you make some earnings on the way. So let's go ahead and click on this and create our first voice. So with voice design, you can either design something completely from scratch or you could provide your own voice for a clone. This is the fastest way, but you do have the option to clone your own voice, remix with an existing voice or try to, you know, make a more professional clone. So here, it actually needs 30 minutes of your audio in order to make a very clean and professional audio clone. But you don't really need to do this if you want something quick for your own use. So let's explore the first one. Voice design. And we get this box where we get to describe the voice. So basically, you can let me extend this a little bit. Talk about the gender, the age, the pitch of the voice, what you need the voice for. Sometimes that helps. Like if you say you want ASMR voice, it's going to make it kind of whispery. So you can explore that as well. And down here, we can get some random prompts to get started. So if I click on this, we're going to get a high pitched female voice. Click again, an older woman with a southern accent, and you can see how the prompts are differing in the length. It doesn't necessarily have to follow one template. You can just put in what you want and get a nice voice to get started. If you click on settings, we can work around with the loudness of the voice and the guidance scale. So the guidance scale is how closely this new voice is going to follow the prompt. You want something to be basically very controlled based on your prompt, you will increase that to high. But if you want 11 labs to be a little creative with how it interprets your prompt, you can keep it on the low side. I would recommend keeping it somewhere in the middle just to get started. And once you have that voice, you can edit it to either side. So instead of using one of these templates, we're just going to describe our prompt, so that way you guys can follow along, and we can all make one voice to get started. So what I'm going to do is try to make a voice for a news anchor. And I'm going for a female, maybe middle aged, confident voice, serious voice suitable for the news. And I'm just going to describe that right here. So a confident middle aged female voice for news anchor purposes, we can say speaks very clearly. So here's my prompt. Very straightforward. I just basically wrote down what I need. Now, on the right side, we can work with the loudness. We don't want her to be too quiet or too loud, so we can go for maybe 60%. And as for the guidance scale, I do want it to bear these key terms in mind. So I'm simply going to increase the guidance scale to 60, as well, because I don't want, like, a very excited news anchor. I want someone who's confident senior and someone who can relate the news very well. So a confident for a senior? Let's put that in. And we can also mention the accent, if you want. So American accent, English. This is optional. You can leave that out and then see what ElevenLabs makes for you. So once we're done, we can see that it will cost us 350 credit. We can generate the voice and then make adjustments after it has been generated. Good evening, everyone. Tonight, we're bringing you the latest developments on the economic outlook, following the Federal Reserve's recent announcement. Our team has been tracking the market's reaction closely. That's the first variant. As you can see, it's really good, suitable for relaying news. Let's look at the other two variants. Good evening, everyone. Tonight, we're bringing you the latest developments on the economic outlook. Good evening, everyone. Tonight, we're bringing you the latest developments on the economic outlook. So I will go with the first one. Good evening, everyone. Just because it's a little louder and a little more clear. Now, the clearing throat part is actually an interesting addition because news anchors are often live, and that may happen, so I could even see what that sounds like. And once I'm happy with it, I could select the voice. But if you felt like it's, you know, too fast, too high pitched, you can make an adjustment here and then generate the voice again. So I'm just going to select voice one. Now we're going to give it a name. So let's call her Maria, use Anchor, and then we could label it. So language is English. This way, it could be used for organizing your voices, but it could also be used when people try to download your voice. So the more labels you have, the easier it is for them to find your voice and download it. So accent we can put what we have gender female. And there we go. I have four labels that will help people find the voice and for my own use. I could change this if I want, but I think the prompt is a good indicator. So let's save this voice. And now it will be added to my library. So with this new voice, we can either generate this speech. So I'm going to type something in. This will be the voice that will be used. I could create an agent with this voice, or I could have it, narrate a story. So audiobooks is something that we're going to look at in the later lessons, but this is how you get to use your voice. These are just a few examples. Of course, there's studio and other stuff down here, even dubbing now when I go to voices, my voices, Maria is right here. So I could use this for text-to-speech, have it narrate a book for me, change the voice if I want to and use it for different purposes. So what we can do is just click on use voice and do a little script right here. Let me just choose Maria first. Going to do V three for a very natural style. I'm going to grab a sample news segment just so she could use the right wording and the right style, and we're just going to paste this right here. So I'm just going to paste that in. We have something very simple. I will not use audio tags just because the voice that we generated should kind of follow that news anchor style anyway. But if you want, you can put in different audio tags like laughter, coughing, you know, whatever you want. But now we're just going to generate that speech. If you want, you can further enhance the paragraph with audio tags, but I'm just going to see what this looks like on its own. Last Friday, students from Sunshine Elementary School planted Oh, so we do have the cafe audio effect from the previous lesson. Let's cut that out and regenerate that. Last Friday, students from Sunshine Elementary School planted 20 young trees in the school garden to keep the environment clean, green and cool. Last Friday, students from Sunshine Elementary School planted 20 young trees in the school garden. So as you can see, that's really good for my application. She's confident. She's very clear with what she says, and you can easily use the voices that you generate using this section. Other things that you could do is clone your own voice, and that's going to be something that we'll look at in the next lesson. But now let's take a look at voice remixing. But let's go into this guy, and we could first add a voice reference. So I'm just going to add Maria right here, and we're going to transform her into a different style. So let's say I want to make her British, and let's actually change the gender as well, so two things at the same time. So turn this voice into middle aged British male. Here we can look on the prompt strength. So how much the prompt alters the original voice. I will keep it medium because that's exactly what I'm doing, the accent and the gender. Could get some inspiration. There's also a guide if you're not sure what to do, and you could even give it a script so you can see the full application. So I'm going to paste the same news segment that we saw earlier. I go to save that right here, and we're going to send this in. Last Friday, students from Sunshine Elementary School planted 20 young trees in the school garden to keep the environment clean, green, and cool. Last Friday, students from Sunshine Elementary School planted 20 young trees Last Friday, students from Sunshine Elementary School planted 20 young trees in the school garden to keep the environment clean. So, these were three samples. I do feel like he seemed a little younger. I was asking for a middle aged British man. So let's say generate more, and I'm gonna increase the prompt strength. So let's actually do it here. Generate you can also use these as reference, combine it with Maria, but I'm first gonna work with the age of the new voice. Sunshine Elementary School planted 20 young trees in the school garden. Last Friday, students from Sunshine Elementary School. Last Friday, students from Sunshine Elementary School planted 20 young trees in the school garden to so this is pretty good. I can now save this as a voice, the same thing. You just put in a name, give it some labels, and a description. Then this will be added into your voice library. And this way, you could have different variations of your first voice, and it's really good for different sorts of content. So say I have one news piece, I want to try a female anchor and a male anchor and then finally make my decision. So this is a pretty easy thing that you could do. You could make another remix right here, give it a different script, maybe choose one of the main voices and see what comes out. Go ahead and give this a try and see what sort of voices you can generate. 21. Voice Clone: The next thing we're going to do is clone our voice. So just as I'm speaking into this mic right now and all of these voices are available, we can actually have our voice, the same voice you're hearing be one of these options. So when I do text-to-speech, I can have it read the text. As myself. Right away in voices, there's two ways to access this. Number one is hitting the plus right here and then instant voice clone, or you could go over to this blue thing. If you click on it, it's the exact same thing. It doesn't matter which way you go. And we already looked at this one, now we're going to go down to this. This is a really simple way to create your own voice. If piece of audio that you don't want to record the whole thing for, you may consider cloning your voice. Now, I will say that this is not exactly the perfect replacement for your own voice recording because as we know with AI, it's inconsistent and if you plan on doing this professionally, you may want to find a better alternative. But nonetheless, it's a cool thing to do. And all it asks from you is a ten second audio. You need to speak for ten second in a microphone, make sure there's no noise around and your voice is crystal clear. Say everything in a slow and clear voice, don't rush anything. And that way the 11 laps can study your voice and then replicate it. All you have to do is hit record audio. Choose the mic and it will do a three second countdown and you get to start reading. I will say a really random sentence. It doesn't have to be the full 10 seconds, that's just the ideal amount. I will start talking about the weather. Lately, it's been really hot and I have planned not to go outside as often. I'd rather sit home and relax all day. So 10 seconds of audio. I did 11, but it's fine. Here is your recording. If you'd like to do another one, you can just hit delete and replace it. Here you can listen to what you sound like. Lately, it's been really hot and I have planned not to go outside as often. I'd rather sit home and relax all day. So there we have it. There is an option for ElevenLabs to remove the background noise. I'm going to disable that because I didn't have any, but if you feel like your audio does, just turn that on. Now it's ready and I just have to click on next. Now I get to put up some information regarding my voice. You can preview the voice right here. Airplanes soar through the sky as autumn leaves drift down. That's how 11 labs interpreted my voice. Let's give it a name. Call it My voice. Then once again, we have to add labels. I'm speaking English. And then we can give it a description. My casual voice. Next, you just have to confirm that this is your voice. We talked about this in the ethics lesson. Do not replicate other people's voice, especially not without their written consent. But this is my voice, so I have all rights over it and I'm just going to click it, save the voice. And there we go. Now we have the option to try out our voice. We have generate speech with text-to-speech, speak with your own clone, where we can create a conversational AI agent, and then we have that dubbing feature using studio. I'm just going to hit Skip for now because we already know where the tools are. There is my voice. You can go to my voices as well, and it should be the first thing, zero days. I could begin using the voice by clicking that icon and start typing something. The model it recommends, you can just use that and then I'll leave everything as default. Let's turn this off and generate the voice. What a wonderful day. We can make it slower, less unstable. And add some exaggeration. What a wonderful day. That's my voice right there. It doesn't I said, sound exactly like my voice as I'm speaking right now. It's just a fun thing you can do. I wouldn't see you guys using this professionally in any way, but I believe the more they work with their models, the better this feature is going to be. That's how you can clone your voice. It's really straightforward. If you wanted to change or delete your voice, you just go to voices, my voices, and then delete. And there we go. Now, we don't have any more clones. Now that we know how to clone and design our own audios, let's go ahead and start a project from start to finish where we take a web page and turn it into a podcast. I will see you guys there. 22. Scary Podcast Part 1: Now we're going to start our very first project. We know how to use the tools. It's now time to put it into one massive audio. With this first project, we're going to do a scary story podcast where we take a web page, turn it into a podcast and then have three people in that podcast. We have a host and two guests. What we're first going to do is find that web page that has that story we're looking for. I'm just going to head over to the Internet and just go to Wikipedia. Usually they have really long and detailed pages regarding a certain incident, a certain object or whatever. Going to look for a natural disaster one because those can be really well converted into different emotions, fear and all that stuff. Let's go to Wikipedia. You can use any other website, just make sure that you have the rights to use it. I will look for a tornado. So right away, we have some examples. This one in 2007. We can look for other ones, even this one. I think I will go with this extremely powerful. I'm just going to make sure there is a lot of information about it. We have comparison to other tornadoes that's good, the aftermath, significance, the damage, so it's perfect. I'm just going to copy this URL, and we're going to go over to Go Studio and import from URL. Let's paste our scary podcast and choose a natural conversational voice since we are going for a podcast. Lovell, trust a few, do wrong to none. We make our own fortunes and we call them fate. Nicht Due to daraufin as. Hi there, and welcome. Grab a coffee, settle in, and relax. And let's chat about this. Gratitude is riches complaint. A single rose can be my garden, a single friend my world. We make our own fortunes and we level, trust a few. I think I'm liking Charlie for now, so I'm just going to click on his name. Right now, we don't even have multiple voices in the URL, so this will not be useful for us. I'm just going to have Charlie be the voice for this podcast first. Once we are happy with the way he sounds and the way the podcast is extracted, you can see right away, it's 17 minutes. If you have a longer text, it's going to be longer. Over here, it's trying to convert this table into something conversational. I'm going to have Chad GBT just take out the numbers and leave out the details for this. As you can see, it removed all of these numbers and only kept something that's more natural. It kept the dates, the number of the miles and all that stuff. You can do this and refine further if you don't want any numbers in this section or you want to take out certain words. I'm just going to copy this and put it in 11 labs right around here. Here it talks about this tornado and then talks about the other four. Okay. You can see that it also has the titles from the Wikipedia page, but we do want to make it sound a little bit more normal. Let's add in a few words. The reason why we're changing from normal text to a heading is because it's going to emphasize it and read it in a way where we're moving to a new section. There's going to be pauses, and normally, it's going to be a little slower than how it reads the other normal text. Now, just like as we learned before, you could add the pauses and the sound effects, but for now just focus on finalizing your text. If you want, you could go back to Chat JBT and ask for more inputs, maybe different reactions to what Charlie is saying. For now, I'm just going to finalize what Charlie says and it generate. While that's generating, let's give this a name. I command A and then generate one more time. Let's talk about the 2007 El tornado and its aftermath. The 2007 El tornado was a small but extremely powerful and erratic tornado that occurred in Canada during the evening hours of Friday, June 22, 2007, powerful Five tornado that struck the town of Eli in the Canadian province of Manitoba, 40 kilometres west of Winnipeg, was known for its unusual path, how it was during its path, its rope to cone structure, as opposed to a wedge structure, and how it is unique. But what were the meteorological synopsis? Okay. Notice how it did pause between the normal text and the headings. I did heading two, but you can do heading one as well. Any of these headings do signify that the voice has to pause and emphasize moving to a different segment of the podcast. Let's start from here since this is where we took it from Taji BT and see how normal it sounds. Were there more? Yes, there were. In addition to the Eli F five tornado, four more tornadoes also affected Canada on June 22 to 23rd. After the historic F five tornado hit Eli, Manitoba, another powerful tornado touched down just 10 miles to the west near Oakville on June 22. Rated F three, this tornado damaged trees, outbuildings, and a couple of grain storage bins along now that the base information is up to my liking, I can now take this back to hat TPT to make it a little bit more human. It's going to add in ms, pauses, maybe some stuttering, anything that would make this like an actual podcast. Because right now it only took the information from this webpage and converted it into voice. But since we are going for a podcast, we want to start with a strong base, and once we have Charlie's parts put in properly, we can add the other two guests. Let's take all of this, then go back to our chat and first describe the scenario, then paste a text. Okay, I'm going to let this run and as you can see, it's assuming that you want this to be the script, but we don't really need that capability right now. When it's done building this, I'm going to have remove these segments so that it's only the text on its own. But it doesn't look like it's going to generate more. Let's see. But right away, you can see that it's adding the words, the three dots, dashes, and it does emphasize on certain words, just like a regular podcast. It's done with the conversion, and as you can see, didn't add anymore of these segments, so we don't need to regenerate anything. Let's copy this body text. And replace it with what we have here. Now we're going to use the different headings on the places where we want Charlie to pause. This one's a beginning definitely like that. Here's another segment. Okay, so we have two segments here not counting the beginning. Let's go ahead and hit Command or Control A, and hit regenerate. Okay, so let's talk about the 2007 Eli tornado and what happened after? This wasn't just any tornado. I mean, this was Canada's first ever official F five tornado. And yeah, it might have been small, but it was ridiculously powerful. And, honestly, just straight up weird. It hit Ely Manitoba, a quiet little town about 40 kilometres west of Winnipeg on the evening of June 22, 2007. Now, here's the creepy part. This tornado didn't follow the usual rules. It had this strange roping shape, not the massive wedge you'd expect from something that strong, and it moved slowly, like weirdly slow. It twisted. Let's move on to this last sentence. The total damage estimated at about $39 million back then, which would be over $56 million today. So what made that day so volatile? Well, meteorologically speaking, the setup was like textbook chaos. A low pressure system moved in from Saskatchewan, and a warm front parked itself north of Eli. And yeah, things exploded. Now, here's what happened on the ground. The tornado touched down just north of the really important to use those heading words, and that's the perfect opportunity for you to add some music in there with a pause or some sound effects. But other than that, you can see how using HGBT was able to add that human touch to it. It does pause at certain times, but feel free to move these words to other places or just remove them altogether. But given that this is a podcast and we're trying to make this sound like Charlie is just your average guy, this works perfectly for our case. So now that we have Charlie's part all done, we're going to move to adding our two other guests. So that's something we would have to do with a different tool. And that will be for the next lesson. 23. Scary Podcast Part 2: So now we have something in our studio which we call depict our Native podcast. Previously we took a Wikipedia page and we converted it into a podcast using JAGPT. Now, you could fully rely on just 11 labs for this for turning that Wikipedia page into a podcast. However, for my case, I did want them to have our host, which is Charlie Charlie to have more pauses, some ms, some s in there just to make it sound more human. But you're totally free to not even use JAGPT and only rely on 11 laps. So now we're going to move from here to create a podcast. Immediately, you can see that we can use existing project. Now, you did also have the option to put your Wikipedia page here. However, I wanted to start from scratch so you can see the progression and how you can put in your own adjustments as we go. Let's go to use Existing Project and choose our tornado podcast. We want to keep it conversational. Let's keep it a little longer because it was 17 minutes. Our host is still Chris, but we do have a guest right now. So let's decide who our guest is. She tore her gaze away from her ruined footwear, still very much grieving the loss. Let's search for conversational There's a Nadal on. If you spend your whole life. So I will go with Jessica. You can see she has that conversational tag on it, so it's perfect. And then the language is in English. If you want to have some sort of focus, you can put in the words. So let's put the word tornado. Let me see. Let's to highlight the dangers of a tornado pause on the numbers, I guess, damage numbers. I just gave ElevenLabs two directions to follow. You can see start with some sort order, highlight, pause, and then tell it what to do exactly. You can add as many of these as you want as we saw earlier or delete the ones that you no longer need. Save the changes and let it generate. I'm going to let this run since it's a little longer. You can see that it generated another project here. So once it's done, I'll be back so we can hear what it sounds. Okay, our podcast is ready. It's 10 minutes, 24 seconds. It did cut down from the 17 minutes, and that's mostly because it shortened the paragraphs. As you can see, there aren't any of those bulky paragraphs that we had before. That just shows how amazing this podcast conversion is within 11 laps. And look over here. I even gave us a cool title, the unexpected F five, a tornado surprising power, even though we had it named the tornado podcast or something. When I hover over them, I can see who's saying what green is Chris and blue is Jessica. So without changing anything, let's hear what these two have to say. Tornados are often measured in miles of destruction. But in 2007, Canada's most powerful twister ever recorded was barely wider than a one lane road, and it rewrote everything we thought we knew about these storms. Well, that's fascinating. How does something so narrow pack that much power? You know what's really mind bending about this? This F five tornado that hit Eli, Manitoba was just 35 yards across at its strongest point, but it literally picked up entire houses and spun them through the air like toys. Hold on. F five. That's the highest rating possible, right? So that was a little sneak peek. I'm not going to have you listen to the entire thing. But if you wanted to skip through, remember that there's a timeline down here that you could just put your plate head over and then skip to maybe Chris or Jessica whenever you wanted to. Here, everything is in the regular text. You can see the attribute right here, but we could change the way they read it by changing some things to headings. So let's see. So here's our transition that we had initially. You can see it kind of changed it. Initially, it was like, but what about the damages? Now they turned it into, so what kind of damage are we talking about here with that kind of focused destruction? So let's change this to heading one and then look for another one. This is about the damage. All right, I'm just going to keep these two as our key transitions and then simply regenerate this part. Well, you're not wrong. At one point, it just parked itself over the town's flour mill for four straight minutes. Just imagine watching a tornado hover in place, methodically destroying everything beneath it. So, what kind of damage are we talking about here? With that kind of focused destruction? Let me paint you a picture, and this is where the numbers get truly staggering. The total damage was estimated at $39 million in $2,007, which would be over $56 million today. But here's the most incredible part. Not a single person died. Wait, what? How is that even possible with an F five? So we did pause with the heading just as it did with the first tool we used. And I just forgot to mention that we did ask ElevenLabs to highlight the damage. So you can see how it has multiple sentences. We had one here. There's another one here. Damage. Let's see. More damage. It's reiterating the damage part of the podcast numerous times. You could go back and change that into something else and really make this podcast your own. I do have all of these words. What I do want to do is maybe add even more pauses between these two headings just so that it's really a full transition. Let's add a pause right here, 1 second, and then another one or one more second. And then another thing that I want to do is do an intro for Chris and Jessica and then bring in our third person. We mentioned having a host and two guest. So let's hit Inter right here. Hello, everyone. I'm your host. Chris. And this is I'll just use the title here. Okay. So let's mention Jessica and then find another voice. Waves crashed. I think Alexis is a good one, so let's just say Jessica and Alexis and then have each of them introduce themselves. All right. So I have four new sentences or lines, should I say, and this just makes it a lot more natural considering we do have three people in this podcast. Now all you have to do is assign each of these sentences to their lines. Alexis is this line actually. When I go over it, I could see that it's not Alexis, so I'm just going to click once on this. Our voices are right over here. We accidentally brought in Archer, but that's fine. Let's turn this into Chris. Apply this one to Jessica. That one's already Alexis, and this is back to Chris. The rest is Chris. Here's Jessica reacting. Let's turn this to Alexis. Just go back and forth until we have a conversation going on. Seeing how Chris is the host, all of these lines where there's a lot of information about the tornadoes, I'm assigning that to Chris. Jessica and Alexis are just reacting to what he says because they are the guest. Let's two of them, Alexis this time and not have a routine transition. Okay, so we have a good back and forth between our subjects, and now we are ready to add some sound effect. When he is introducing the podcast, we do want an audio to play before it, and maybe at the headings, put some sort of sound effect, maybe a commercial break in the middle. Something that will make this work like a podcast. We're going to do that in the next lesson. Just make sure that by the end of this lesson, you are happy with the words that each of the guests have. You could even add more people to this podcast. I'm just going to keep it at three. But go over each of the words, make sure that they do belong to that person you want. And you don't have a random person in your podcast. When I removed Archer, you can see that he's no longer here and I only have three people. If you see another voice here, it means that somewhere in your podcast, you did assign that voice to a certain text. Down here, again, you can see a colorful transition. If you saw that there's way too many greens back to back, then that's assigned to switch it out with a different voice. Let's move on to the next lesson where we add our audio and just finish up this scary podcast. 24. Scary Podcast Part 3: I let's finalize our scary podcast and add some sound effects. Previously, we took a Wikipedia page, converted it into a conversational podcast, then brought it in here where we added two other voices. All I did right now was put the pauses after the headings, but you can add more pauses wherever you want. Even between the sentences, you're free to do that. Now, what I'm going to do is make an intro for this podcast. So right before hello everyone, I'm going to put in the sound effect. Let's do a tornado Okay. I like the first one more duration wise, let's just do a shorter one. Hello everyone. Hit apply. Then after everyone, I'm going to add a pause, it's the intro essentially. Then over here, we can add, let's see, either a drum roll or a next page sound effect. Let's just do that. To page flip. Okay. And in terms of the duration, once again, I'm going to do a little bit of it and not the entire sentence. I like number three more and just be sure to keep this on the low side since we don't want it to go over our Cris basically. All right. And then what we're going to do is just adjust the voices a little bit. Jessica's is a little bit too exaggerated, so I'm going to lower the stability. Sorry, increase the stability and just make her tone a little slower. It's safe. Alexis was fine, so there's no need to do anything with her. Chris, I do want his voice to be a little slower as well. S save and I'm going to find other places to add a pause. Okay, so we added our sound effect. What I'm going to do now is just have a small section of this generated so you guys don't have to listen to the entire thing. Let's do this one part where we had the sound effect, the pause, and of course, we can see how the voices sound like after we made our adjustments. Hello, everyone. I'm your host Chris, and this is the unexpected F five, the podcast that reports on the world's most dangerous F five tornadoes. I'm here with our two guests, Jessica and Alexis. Hey, guys, I'm Jessica, and I'm so happy to be here, and I'm Alexis. So excited to talk to you guys about these spooky events. Alright, let's get into it, then, guys. Starting with what we know about the tornadoes. Tornadoes are often. Alright, then that sounds a lot better. I do feel like I want to change Alexis. It seems a little bit too robotic. So what I will do is go to this part or part and find another voice, basically. Let's look for conversational and find a female voice that sounds, you know, a lot more normal. In the ocean of competition, be a swift fish. The Ocean heights treasure. He up yours. Just trust yourself. There is no greater. I think Area sounds good here, so let's switch Alexis out for Area. Then let's see. All we have to do is just grab this part and click on Apply with Area. And then repeat for the rest of the thing. I'm going to try another generation for you guys since we added a page flip here. It starts with Chris. Let's apply Area over here. Once again, I do want a slow and steady voice. Let's do that, it safe, and then do it here. I think I'll change this to Jessica so there's some variety. All right, let's grab this whole part, generate, and listen to another segment. What kind of damage are we talking about here? With that kind of focused destruction, let me paint you a picture. And this is where the numbers get truly staggering. The total damage was estimated at $39 million in $2,007, which would be over $56 million today. But here's the most incredible part. Not a single person died. Wait. What? How is that even possible with an F five? You know, that's the real story here. It's about preparation meeting. But before we get into that, let me break down exactly what made this day so perfect for tornado formation because it's fascinating from a meteorological standpoint. Hmm. I'm guessing multiple factors had to come together just right. Oh, you better believe it. We had this low pressure system moving in from Saskatchewan temperatures in the high 20s Celsius. That's around 80 Fahrenheit. And humidity levels that would make a rainforest feel dry. Okay, and that is sounding a lot better. The page flip I do feel like has to be moved a little bit, so I could just grab it down here and move it till the end, maybe here. After she says the word destruction, page flip. Then I would just continue altering the rest of my podcast, being sure to add the pauses where it needs it, and then adjusting my voices further if they're not reading something right. But for my case, they read everything fine. I didn't need to do much adjustments over my scary podcast. And now we're basically done. We took a Wikipedia page and turned it into a podcast. What I could do is just export it as a single file, p three, and then just listen to it when I want to. This is also a nice way to take articles that are way too long and then just turn them into a podcast so you can listen to it when you're walking around or you're in transport. That concludes our first project. The next project, we're going to try to do a different type of voice generation using the same tools, but application wise, it'll be for something other than a podcast. I will see you guys there. 25. Audiobook Part 1: Now we're going to move on to a different project, which will be to make an audio book from scratch. What we're first going to do is get that book that we're trying to turn into an audiobook. I'm starting here in ChachPT, but this would be assuming if you already have a book that you have the right to turn into an audiobook or that you have your own stories. Just for practice sake and for you guys to have some hands on experience, we're going to use this tool to make a book out of certain words and prompts. Open up Chat GPT and first decide what your book is going to be about. You have to know the subject, what type of book it is, how long it is, and who is like the characters. One thing that you could do if you don't know where to start is have Chat JBT give you a few ideas as to what you should turn into a book in the beginning. For example, if you have a YouTube channel where you talk a lot about real estate, you can come to Chat GPT and give a prompt like this. I enter, and then we're going to get some suggestions. From there, we get to continue on. What's interesting is that Chat Tu But will give you the title and then it tells us why it works and some details inside that book. We got what they don't teach you in school from rent to Riches, civilized, real estate Diary, and all of that stuff. What you could also do is combine these ideas into one big book or maybe just take out a few of these bullet points and replace them with something else. Let's see. I think I like the House hunter survival guide. That name is pretty good to me. I will say I like number one. However, remove the quiz part. And replace it with let's see what we can replace it with the quote real estate jargon. Here I just did a little switch and now it's giving me more details. Who are the audience? For some buyers, renters, real estate browsers. If you already have your channel, just double check that this is indeed your audience. If it's not, you can switch it out. For example, I will say my audience include middle aged to young adults. Age range, I can put 50, maybe 25 to 40, let's see. Then this is just a random age range. Those who live in California, Los Angeles, California, those who have no kids and dogs. I'm getting more specific regarding this book. This is our audience and let's see. Maybe we can add one more thing. And work in the fashion industry. Here's a cool snippet that it does, the tone and vibe. Think a stylish friend who already bought a place and is walking you through it over Macha. Sharp but casual language, real talk about money, references to fashion design and LA specific quirks. Then this is the chapter. There are ten chapters, which is good. And now that it looks like this is going to be a good book, I'm going to specify from what point of view this is written from or if there's no characters at all. Then lastly, how long is this book? Let's say, in terms of who whose point of view this book is from, I think this person, sharp a casual language is good. But we can make the narrator use this one. But make them a well informed real estate agent who knows a lot about this particular audience. Then give me only three chapters, totaling to 50 pages. Make this a quick and fun read for my audience. So this will be a whole different page and here Armour chapters. As you can see, it's giving me bullet points and you could keep that idea. You can see it's giving us a snippet, summary or something. What I want to do is convert these bullet points into paragraphs. Let's close this real quick, come back here and ask for a full version. This is good. Give me the full version in paragraphs. Then I'll cut it down even more. So I'm going to let this run and then I'll be back to show you what our script looks like. Then if we need to do any changes, we get to do that before we move to ElevenLabs. Okay. So here is my book right now. It's not that long, which is perfect. But as you can see, it went from bullet points to actual paragraphs. Now, keep in mind that you do want this to be a audiobook, so make sure yours has the indications for chapters and anything else that you want to put in. You could have bullet points, but I would recommend putting it in between paragraphs so that it's a more natural audio book. And that is the name of our book. In the next lesson, we're going to put this into ElevenLabs and start matching that audio to fit perfectly with this tone and for our audience who are people that are looking for a nobs guy to buy their first. Let's move on to the next lesson. 26. Audiobook Part 2: Alright. Now that we have our book, I'm simply going to copy everything from here. I think this entire thing is the book, so you can just hit copy right here. Go to studio and create an audiobook. So uploading a document is optional. We can also copy paste. So let's not do anything here, but go over which voice we want to use. So let's look for narratives and see what we have. She tore her gaze away from her ruined God has given you one fact. Let my voice become your voice. The people who I have crossed. Let my voice be. So, like, I guess, for me, the main thing is, like, I don't understand why people have to get upset. It's like Theresa She tore her gait. Net. I'm Sophie Lang. Hi there and welcome. Grab a coffee, settle in. I will go with this guy that we didn't use in the last project just to give him a chance, and then we can hit Create. So we're brought onto this page that you could also go in with the studio, but we're simply going to paste what we got from Chat GPT in this box. Then notice here we have plus chapter. We have three chapters. So we're going to hit Plus chapter here and copy the Chapter two and beyond. Delete it from this one, go to Chapter two, paste it, and then copy Chapter three over to another chapter. There we go. We have Chapter one, two, three, and I'm going to match the names over here with what we have over here. Go to the three little dots, rename, paste it with Command or Control V, click away, and repeat this process. We got our chapters all done. I'm going to remove some of the things that I don't need this guy, instead I'll call it preface. All right. This is the title of our book. I'm going to make that heading one so that it pauses and registers this as a separate thing. In terms of our voice, let's lower the speed and keep it more stable and more professional. I'll lower the audio a little bit. It's safe, and let's generate. Let's start by generating a small bits and then make our changes. I'll grab all of these guys, then it generate. The House Hunters survival guide, LA edition. A OBS guide to buying your first home in Los Angeles without losing your cool, dog or sanity. Preface. Hey, I'm not your mum, your financial advisor or that guy on YouTube who thinks every home should have a panic. Okay, so right away, I noticed that I want this line to come right after this. So let's just delete it and bring it up. Colin, put it here. And I think I'll keep A and change this to maybe disclaimer. Then since it's a male, let's do Tad. Then what I'm going to do is do one more change, increase the style exaggeration, and the audio is a little too low. Let's bring the playhead back and redo this part. The House Hunters survival guide, LA edition, ANBS guide to buying your first home in Los Angeles without losing your cool dog or sanity. Disclaimer. Hey, I'm not your dad, your financial advisor or that guy on YouTube, who thinks every home should have a panic room. I'm a real estate agent who's helped people just like you stylish, dog owning creatives, juggling a dream and a deadline, buy their first place in LA without having a total meltdown. I'm here to make this whole process feel way less overwhelming, a little more empowering. And honestly, kind of fun. Right, so that's a lot better. And now we can move on to how it reads the different chapters. So I'm gonna continue pressing play, so we can see how that sounds. Chapter one, welcome to the LA circus AKA the Market. So for the title, there should be a break here. That's what I'm going to do. A five second break, half a second. And then here, I'll do the same thing. Okay, let's hear it one more time. Chapter one, welcome to the LA circus AKA the market. Let's get something straight. Buying a home in Los Angeles is not for the faint of heart. This is the city where a two bedroom craftsman can cost more than your entire college education, and you'll still need to beg someone for street parking. Competition is wild. Cash buyers are everywhere, and the market changes faster than your favorite pop up coffee shop closing for renovations. Right and that's pretty good. So what you're going to do is continue building your chapters if you'd want to. But honestly, because ElevenLabs has a built in audio tool, there isn't much for you to do here. So you just declare your chapters, add a pause whenever you want to and adjust your voice. Now for the title in the chapter titles, I'm going to do the same thing where we added a pause. Let me just go over here. This is heading three. Keep things consistent so that your voice can read it the same. Let's add another pause here. I will do maybe 0.4 seconds, and then the last one. One more time, heading three, another pause. Now that I have all of my pauses and all my text, I can go ahead and export this and perhaps in a different software pair it with a music if you want to or add sound effects as we already learned how to do. However, I'm going to keep it plain because that's just how I prefer my audio books, but you're free to do whatever you want. Because you know how to use each and every one of these tools. That's how we can make an audiobook. You started from HAGPT finding book ideas. We tailored those ideas into a very specific niche of an audience. Once we had that, we got some bullet points. We turned those points into paragraphs, limited that to 25 pages, and then we brought it into 11 labs where we declare the chapters and started the translation. Now Archer is reading our book for the audience. I hope you guys enjoy this short project. Let's move on to the next lesson. 27. ElevenReader: Once you have a story file, you can easily turn it into an audiobook using ElevenLabs. We saw how to do that, but there has been an update, which is the ElevenReader. So basically, you can upload your eBook, and it will start narrating it for you and you can earn from that narration. So previously this is where we went. We would upload our PDF file or EPAP to the narration style chooser model. But now when you go down here, you can upload the same story and just have it go to this platform. So I have a sample story in dot TxDFmat with two characters. Going to upload that right here. I just uploaded that from my computer, and we're just going to click on create and preview book. So here's my story. It's, you know, written line by line, and we're getting our different characters. The first one is, I could, give it a name since we have two characters, we can either have it be multiple characters, narrators, or just one. So that's up to you. But if you click on either one, you can assign who is speaking. Here I have some tags about the characters. I don't really need this to be set anywhere. So we're just going to delete that using the backspace key. So you can easily make adjustments in this page. And the same thing here. That's the name of our story. Scroll down. Everything else looks good. Now, what we could do after that is play around with the volume. This is pretty similar to the studio layout, but we're just getting that eBook capability. So if you want, you can add in some music from here, make your own or drag one in. You can make different characters. So this is the current narrator for a story. The lantern at the edge of the sea. Mara found the lantern on a rainy Tuesday. So this right here is not something I want. I want something a little deeper. So I'm just gonna go here and change the voice. So let's go for Explore. We will go for English, see if there's an audiobook option. I want to get a preview first. Ne IguodaTokEviacos, OEtoato Meg ir. Let's just make sure it's English. The winds of the lands between whisper of your deeds. Do not be afraid to ask largely. So what I'm gonna do is click on Marcus. That's the voice that I want, and it's going to basically reassign the narration from Eric to Marcus. So there we go. Everything is now blue, and we could do a little preview here. Mara found the lantern on a rainy Tuesday, half buried beneath the wooden steps of the Old Harbor Lighthouse. She had come there looking for shelter from the storm, not treasure. But the small brass lamp caught her attention. So once I'm happy with the story, the speech, the type of voice, I can now click on Export. Can export it as we saw either as an audio or a different file structure, but I want to point to ElevenReader now. So the cool thing about this is that you can earn from this feature. You get 0.25/hour, that is streamed and 60 per direct sale. Now, this is a sample story, but we're going to try it out anyway. You can do dynamic narration, so they could kind of people that download your story can choose their own voices, and this way, there's more flexibility on their end, but you can also just publish what you have and not give them that option. So I'm just going to do original audio for this. We can click on Publish. All right. So once we come here, we can give some more information about the storybook. We can create our author profile that's us. Upload your picture, give yourself a name, the language, of your profile, and if you have a portfolio, you can put the URL as well as a bio. I'm just going to do some random stuff right now. Say my name is Megan. You don't have to fill out the rest. Going to the next one, we can talk about the book itself, the story. That's going to be the metadata. About the category, the genre, and all of that. And on the right side, we can see the story being uploaded. So while this is loading, I'll just go here and you can share this QR code with people or download the app yourself. But going back, we can now put in a name, a cover image, subtitles, our own author profile description of the book, and it's pretty much like any other eBook platform. The only difference is that you made the voice completely with ElevenLabs. This is another output that you could do from your creations here. I'm not going to go through the full thing because it's going to be like payout details, distribution details, and this, like I said, is a sample story. But if you are interested, you can take your audio books a step further and upload them on ElevenReader. There's also other options if you want to publish to Spotify. There's in Audio, the video and Audio Native. So these are pretty cool things, and you can explore them further. 28. Advertisement Part 1: For our last project, we're going to include some things that are outside of ElevenLabs and see how we can combine them into one video. What we're going to do is get some footage from the Internet. This could be made with another AI tool such as runway or Mid Journey, or you could use copyright free videos from Pexels, combine them, and then make a dubbed footage, adubb audio for your footage. First, let's go ahead and decide on what the video is about. I'm going to head over to Pexels video. Right over here. And while we're here, let's get an audio tab as well. I'm holding down Command to open a new tab just so we can have some music in there as well. So I meant Pixabe for audio and then pixels for video. So I think what I will do is an advertisement where first we make it in English. We have some nature shots maybe and then have an upbeat music in the background. Once we have our English audio, we will translate that into a different language so that it's like a multilingual commercial. Let's get some nature shots or maybe tourism. This will be a touristy ad, and there's a ton of videos here for free for you to use. Just be sure that if you plan on publishing it anywhere, you give credit to the creator. All the people who made these videos are at the bottom left. Let's download a few of them. I'm going to go in and get a less heavy file. There's our first one. This is our second one. Let's do an island shot and try to get different angles because that's what commercials usually do. Some birds over here, Okay. I got a bunch of videos. I'm going to head over to Chat GPT to get a script real quick. Feel free to do your own script. I'm just doing this for the purposes of this course. Let's say I want to make a hoops 60, let's say 30/32 ad about traveling. Help me write a script. And then let's say who is this for and what should the script I want to emphasize be who you are anywhere and being free. Use inspirational and professional tone that motivates the audience to start traveling. Give a shout out at the end to example channel. Let's assume example channel is my YouTube channel. Very short and we got a couple sentences. I'm just going to have ChachBT remove these snippets and just give me the voiceover script. Very short, very straightforward, and this works well with our tourism shots. That's exactly what I'm going to do. Let's copy this and then go into dubbing, create a dub example channel, source language in English, and let's say the other one is French. French or Spanish, any other language you want. Now, the video source, this is where we're going to have to combine our videos and then put them into one file. You can use any software you want for that. You're just basically taking the videos that you downloaded, sticking them together, and then exporting them for this part. Whichever source you choose, you still need that file. What I'm going to do is go to Premiere Pro and put these four clips together, and then I'll be right back so we can upload them here. Put the clips together. I came up to 35 seconds, but I did that just in case I wanted to stretch out one of the dubs or add some dramatic music. As you can see, it's just a very simple compilation of these videos, no transitions or anything. But one thing to keep in mind is keeping the source clips similar. They're all slow pans of touristy destinations and because I'm talking about individual freedom and being who you are, I do have people walking in different places. Then last part where I'm subscribe to our channel, it's a long shot of this deep. Later, I could put text and stuff. Now that I have my base clip, before I go into dubbing, I'm actually going to get some background music. I forgot to do that. You can use Pixaba. It has free, no copyright music. Again, if you plan on posting this anywhere, be sure to attribute the artists that you can see right here. Let's look for commercial or let's look here. Going to go with piano. This audio works for me. You can also look for other things like travel music, log music, anything else that you want. And now that I have my base stuff, I'm going to go to dubbing and start filling these in. You could put the background music on the video file itself, so you're only uploading one thing, but I'm going to go to manual so I could perhaps have option over the layers. Now that I have my source files, the video file and the audio file, we can go ahead and start making the dub. I will see you guys in the next lesson where we do just that and finalize our example channel advertisement. 29. Advertisement Part 2: Welcome back. In the last lesson, we came up with an ad, and now it's time to actually create the text, the text to voice, and then do a Spanish tub or another language stub on top of it. Now, before I get into that, I do want to try making a music for this ad. We did download that soft piano music from Pixabe in the previous lesson, but you also have the option to generate music right over here. So let's open that in a new tab. So you have variance, we just like the different versions you get to create. This is the duration. I will set it to 30 seconds because that's my ad, but you can put it on auto or any other values. These are your recent projects, but all you have to do is describe your song. Let's do a inspirational let's try something like this and we can also then change it up if it's not to our liking. I set it to one variant and you can see that I'm only getting one right now and essentially we're getting the different parts made for us. 30 seconds like we asked for. And this is the styles that excludes styles, you know, things that we did not ask for. We asked for inspirational, piano cinematic, but we obviously did not ask for a rock guitar or anything like that. Then we got the different sections. Let's hear our first variant and see what we're dealing with. That was actually pretty good. Notice here, it's interesting. I added a sample text for us to better imagine that voice over over this intro. For me, this is perfect, but if you wanted to change it up, you can either continue this conversation, which means adding a new section, maybe you want something else, we hit plus and you add a new section, upbeat remix section or something, then we can once again do the include and exclude parts. So that's how you get to change up your song composition. Once you're done, I'm just going to delete this extra section. You can hit Download or share it with other people. So when I hit Share, it's kind of like a ElevenLabs audio library. You can put the song title, give it a little color, and then we can generate the link to this song, preview it and Export. So when you hit Export, it actually exports this animation. I just downloaded it. Let's see what we got. So that's pretty cool. I just made all of that with AI. Okay, that's my music for the ad. Feel free to still use the music you downloaded from Pixabay, combine them. But I just wanted to show you guys that this product exists. Now we can go to text-to-speech and paste our advertisement. Okay, so I went back to Chachi BT and I got our script. Let's go ahead and choose a voice that's somewhat deep, very slow tone, confident, and just like those classic commercial travel commercials you see on the TV. So let's look for deep mail, I guess. Life is like water. This is your deep. It was one of those rare smiles with a quality of eternal reassurance in it. Welcome to my demo. A voice that doesn't just speak. Hey there. This is Dante. With a deep, easy going tone. The world of artificial. Okay, I think I really like. Hey there. This is with a deep Angelo. And then we had Hey, guys. Life is like So it was one of those Seko, I think, that's how you say his name. Let's try both of them and see which one works better. Let's click on add to My voices and just generate first to see how it deals with it. I will make this a little slower because he did speak a little fast when I did the preview, and then we can decide. There's a version of you that's waiting to be unlocked out there. When you travel, you don't escape who you are. You become more of it. Okay, so that is not the sound I'm going for. Okay, so this is an example of when you could actually design your own voice instead of searching through the thousands of available voices. So what I'm going to do is head over to home, which I have it opened right here and just go to voice Design. So let's just see if we have something to start will be trailer voice. Let's say dramatic male voice. And then here what we have. In a world on the brink of chaos, one hero will rise. Prepare yourself for a story of epic In a world on the brink of chaos. In a world on the brink. Okay, so this is giving a little bit more, you know, thriller and action. So I'm actually going to copy our script and put it into texture preview and give this a different thing. So dramatic male voice say deep male voice. Inspiration and Travel advertisements. Let's try that. There's a version of you that's waiting to be unlocked out there. When you travel, you don't escape who There's a version of you that's There's a version of you that let's try something else. Let's say, slow and confident male voice. So let's go to Chat GBT and have it basically describe that classic travel ad voice. The ones that are, like, you know, slow, they're inspirational, deep, confident. But I just got to put that sound into words. So let's say I'm trying to design a voice with 11 laps, ElevenLabs for this travel commercial. This is what I have right now. Help me get it closer to the classic travel commercials. So this is perfect, national geographic. That's exactly what I'm trying to get to. Now you can see it took this one sentence and built it up into more. Male voice, slow pace and confident, warm and rich tone, and I'm just going to copy this. If we saw that it's not getting us exactly what we need, we can just take away from the prompt little by little. But for now, let's see what the direct result of using hachPT is. And recall we do have the settings for you to either put in a seat, work with the loudness and all that stuff. I'm just going to keep everything at default. There's a version of you that's waiting to be unlocked out there. When you travel, there's a version of you that's waiting to be unlocked. There's a version of you. Okay, so it's definitely a lot better, but it is a little bit too slow. So let's try to fix that up over here, and I'm actually going to add a British voice here. British male voice, slow paced, arm enrich tone. Remove this line. So let's remove the articulation part. And see what we got. There's a version of you that's waiting to be unlocked out there. When you travel. There's a version of you that's waiting to be un There's a version of you that's waiting to be unlocked out there. Okay, so this one is really good, but I just want to make this a little deeper. So British deep male voice. Just in terms of the pitch, I'm gonna let's try actually generating this one first, and then we'll make another one. So let's say travel at voice one. Language is English accent is British. Save that voice. I will go to a different window while that's happening. So let's just replace this prompt with our deep male voice and then put in our script. There's a version of you that's waiting to be unlocked out there. There's a version of you that's waiting to there's a version of you that's waiting to There's a version of you Okay, these three are really good. I'm just gonna go back and listen to our music so I can kind of compare and decide. Let's open it right over here. This is the video. There's a version of you that's waiting to be unlocked. There's a version of you that's waiting. There's a version of you that's waiting. There's a version of you that's waiting to be unlocked. Okay, I think I'll go with voice. There's a version of you let's select it and call it Travel ad voice two, British. And there we have it. So now we have our music and our very own voice. Previously, we made our video, and now it's time to put it all together and then do our translation to Spanish or any other language you choose. So I will see if it's in the next lesson where we do just that. 30. Advertisement Part 3: Okay, let's go ahead and translate our video. Before this lesson, I just combined my base clip and the audio itself, not the music. And this is essentially what I have. There's a version of you that's waiting to be unlocked. Out there. You unlearn, you breathe these too. So the reason why I did that first is because ElevenLabs will have an easier time translating, and this will save you the hassle of uploading the video and trying to match the word that you want at that exact minute. You could go ahead and skip this and just upload your audio first in dubbing. So when you go here, when you want to create a dub, you're able to upload the audio or video source. So I'm doing video source because it has the audio attached to it, but you could just do audio and then add the video later on. I suggest always starting with a video source. Let's call this example channel from English to whichever language you prefer, and I'm going to upload that very same video. There it is. It has the idea once again. Let's create a dubbing project. And if you want to save usage, you can have the ElevenLabs watermark, but I'm just going to leave it be if you wanted to experiment and you're not sure about your video, just check this and it will save you a lot of credits. Number of speakers, we only have one and then we want the full range to be dubbed, and that's it. Let's hit Create. It's going to upload this video and begin the translation from English to French. All right. So I'm going to wait for this to happen. It shouldn't take too long, and then I'll be right back so we can continue. It's finished processing. I'm just going to open it and see our translations. Right now, we're familiar with this interface. You have the original, which is English, and then you have the French. So when you go to French, you can see how it translated the English version into here. Again, you're able to edit this if you know French and it's wrong or if it transcribed the English words in the wrong way. I'm just going to double check this. Okay, so not a lot of texts to deal with, but it's all correct, at least. So let's go to original and try to add our music. So all you have to do is put it right over here for foreground. You can upload an audio right over here, so bring the music. There it is. You can see it's even telling us to import something without a speaker because it won't transcribe that. There is my music. I could rename this, click on the name, call this Music. This is an imported audio. If this original video had some music, it would have showed up here in foreground. But since ours didn't have any sort of music, that's why it's empty. Let's go ahead and play this and see if we want to move anything around. There's a version of you that's waiting to be unlocked out there when you travel. Escape. Right away, I could tell that the music is way too loud and just visually looking at it, the background video ends at this time, but the speaker and audio end a lot sooner. Now, with the speaker part, it's not a big deal because you want some natural pauses in there and when they tell you to subscribe to example channel, you do want there to be a little bit of just nothing until the end of the video, so it ends in a smooth way. But for the music, we could slow this down by grabbing the edge here and matching it to our video. Doing this, it's going to lower the pitch of whatever audio you do this too, so keep that in mind. But because I have some piano and some violin, that's not going to be a problem for me. Of course, I don't want to go overboard. Let me just show you what that sounds like here without the speaker. You can see the music is still pretty much the same. It's just that the pitch is different. But if I overdo this or if I go the other way, I squeeze it, it's still music. It's just very different. So keep that in mind. When you do that, I'm going to hit Command or Control Z to undo. Now, the music is also a little bit too loud. Travel, you don't escape. We want the speaker's voice to be the main part. So just click on that audio and reduce the volume. You you play around with it until it's perfect. Become more of it. You learn, you unlearn, you breathe easier. B who you are anywhere. That's a lot better. Now I have my English speaker and my music. It's time to add some sound effect. With commercials, you want to have the viewers imagine themselves in that setting. Since we are talking about travel, we have some scenery of putter balloons, that could sea, all that stuff, we could add and generate sound effects to put in between these clips. So let's scroll down right here. You can see at Sex track. When you click it once, it's going to just make the track and this screen bar lets you generate those sound effects. Let me first see where I want the sound effects to be. For the first part, I can't think of a sound effect I want here. We could do people chatting in the back, but because they're not in the frame, that's going to be a little weird. But for the second clip, we have the C, and you can put in some SGLs fine by or the waves crashing, something like that. Here again, we don't have much to do, but this we could do something like a crowded area, so this person is taking that freedom and walking through a bizarre or something, and here we can do some wave crash. Maybe we do Segal for this one and waves for this one. Let's go ahead and bring our playhead right where we want this audio to start. We could move the SFX after we've built it too. Click once, and I'm going to decide on the duration pretty long, this long, stretch it over. No audio right now. We didn't put anything here. Don't worry about that. Let's skip over to the other part that we wanted an audio, start one here. I'm trying to overlap the clips because that makes it a lot more natural. Now here, when there's two SFX back to back, so pack light, dream. So we have one of the person walking and then we have one with the beach. So instead of putting them right after each other, you could also layer them. Essentially, we scroll down and we add another Saffx track and we make our second one down here. The reason why we're doing this is because when this guy is fading out, this guy could fade in and that creates a very natural seamless transition. Let's make sure they overlap for a couple of seconds like that, and I have three Aafcs to work with. Let's go to the first one, and I will do seagulls flying by near a beach. You'll be flying by distance. Fade in and then fade out. Let's try something really simple. Avon, you don't escape who you are. You become more of it. You learn. There we go. Pretty good. Siegel's flying in the back. I'm just going to mute just highlight this track so click on this headphone icon so we can just hear the SFX. Notice how good that fade in Fade Out was. If I remove that, I'm just going to copy this. Just get rid of this part, generate it one more time. You can see that's not as good. It just suddenly starts, but by adding that fade in fade out, it's going to be more natural. Fade in, audio and then fade out. Just leave that out. Seagulls flying by in a distance. I don't think we need to put the sound of the ocean because it's really far away. We'll reserve that for this audio. All right. This one, I'll leave it as the volume it has right now. Later, we can listen to the entire thing and then decide or how loud we want the audio to be. Let's move on to the second one where we have this person walking through, let's assume a bizarre, a crowded one. Click on this can do a crowded bizarre with people. Talking in the distance, sellers yelling their prices, I guess. We can experiment with this. I will do one without fade and fade out. Now, the reason why I keep doing in the distance is because I want a faded audio and sometimes the volume alone is not going to be that helpful for you. If you do one without the in the distance prompt, it will be way too clear and way too loud, but that's not the point here. We don't want the SFX to be the main character in this advertisement, so that's why I'm adding in the distance for all of these SFC. If you had the sound really close up, for example, super close up shot of a bird, then you may want to not use in distance. Just gauge what distance there is between us as the viewer and that sound that's in the video. This gave us a shorter audio. If you change prompt, you can regenerate that audio or click on this refresh icon to get a different one. We can just click on this right now, right click and do a fixed duration one. I meant to make this like this first. Fixed duration. Now it meets that requirement that we wanted. The words that the people are saying don't make any sense, but that's okay because it's a background noise. Now, by default, when you generate here, it's doing the dynamic duration. If you did what we did here and you saw that it keeps giving you a shorter version, don't try to stretch it out because that just slows down the clip. But instead just right click and do fixed duration. Now, our last scene is, let me see if this fades in and fades out. Okay, doesn't. So let's do Let's go here. Do fade in and fade out all here. Generated one more time. Meant to do it the other way. So this is an example, fixed duration. It's gonna use the same prompt, but keep it the same length that you want it. A Okay, so that's good. Didn't really fade in and fade out like the Segel one, but I think it'll be fine. For the last one, let's do the sound of the waves. Go in this one down here and do waves crashing. We just do waves at the shore. Got slow at the beginning. Okay, right click. Fixed duration Audio. Hi. Turn it on like that. Okay. I think this will be fine. I don't want two Segel sound effects in one click. So when I play everything together, I just uncheck this so everything is now being considered. Where That's the alluding. You can see that there is no balance between the different audios. So let's go to the first sound effect. When you travel, you don't escape who you are. Click on the first sound effect and just lower that audio. You travel, you don't escape who you are. You become more of it. You learn, you are a little lower than that. Let's try the one here. The bizarre one, I'll go a little bit lower just because it's the labs one and then the beach, we can go 16 decibels, negative 16. B who you are anywhere. That's the real freedom. So pack light, dream big, and start your journey today. Subscribe to example channel. So notice how that transition was really smooth from that bazaar to, um, the beach. And that's because of this overlapping thing that we did. Now, to make this one a little smoother, because you can clearly tell when it starts and when it ends. You travel, you don't escape who you are. You become more of it. I think I will just extend it to maybe here. So it's less noticeable. Right click, fixed duration. And because we're so focused on this audio ending, we could pay less attention to the SFX. You don't escape who you are. You become more of it. You learn, you unlearn. It's a little too low. So let's increase that. Become more of it. You learn, you unlearn. You breathe easier. There. When you travel, it. You learn, you unlearn, you breathe easier. Okay, that's pretty good. I think I'll do another overlapping thing. So let's move this to its own layer as a vex track, and I will just put my playhead to get some direction here, put it like that, and now give it a listen. Make it a little louder. You become more of it. You learn, you unlearn, you breathe easier. Be who you are anywhere. Okay. So that's pretty good. Let's just extend the SQLs to the beginning, maybe a little bit further back, something like that, generated again. So it's like this ends, this starts, then this starts kind of like a go situation. When you travel, you don't escape who you are. You become more of it. You learn, you unlearn, you breathe easier. Be who you are anywhere. That's the real. Okay, so I'm pretty happy with this original clip, and I'll just play it for you guys as it is so you can know what we're dealing with before and then what the French version is going to be. For the French version, there's not much you can do. You can see that it just duplicates the same audio track and simply translates it. So you're not doing much editing here. Let's start over and see what we got. There's a version of you that's waiting to be unlocked out there. When you travel, you don't escape who you are. You become more of it. You learn, you unlearn. You breathe easier. Be who you are anywhere. That's the real freedom. So pack light, dream big, and start your journey today, subscribe to example channel and see where the world takes you. So that's the original version. It's all good. What does the French version sound like? Now, when I play this, you may hear, like a duplicate voice. Iliana. So if this happens to you, don't panic. Right now, I'm in the French version, but we can hear something like the English version in the background. Now, right now, I have my French speaker, my original, and the background. So there's three speakers in this clip right now, and when we listen to it, it sounds a little weird. We have the French speaker and someone's muffled in the background. Don't panic if this is happening to you. This only happened because we had the speaker as part of the original clip. Even though the speaker one original is grayed out, it's still reading it weird because we have the background music. If I just highlight this you don't who you are. You can see that's where the muffled sound comes from. All you really have to do is delete it, and then it should be back to normal. That's a simple fix. Now to export this, you can just go down here, click on Export. These are the stuff that you've done before, but down here, you can decide in which of the languages you want to export and then as what. Pfore video will have the video, the sound effects, the speaker, and all of that stuff. You can even export this as a zip file every layer is its own audiora you can do the timeline data. If you have any captions, we don't audio in this format, in this format or this one. Then when you're done, you can hit Export. It's going to load it up here. You get time stamps and in what language, then you have the option to do a preview and then download it when you're ready. Tvantage oye Kivu said, part two, Seliberte. Hello, Voyage grand. He come and save a voyage shows. I believe example, channel, I voye will amount for salmon. That's how you can make a very simple advertisement with ElevenLabs. We started everything from scratch, from the script, the voice, the audio, speakers, sound effects, and all that stuff. The only thing that we didn't make was the videos that we got from pexels.com. But if you want this to be a full AI experience, there are some tools out there that can make AI videos for you. This concludes our third and final project. I hope you guys were able to get the same result, if not similar to what we have right here. Sure to try even more projects and practice the tools that we learned about. Now let's move on to the next chapter. 31. Creator Tools Introduction: You can now generate image and video within 11 labs. Apart from generating those assets, you can also upscale them, create your avatars, and really have a full production on one platform. This is a new edition, image and video, and essentially, this is what the page looks like. We're first greeted with avatars, which is a way that you can keep consistent characters. So say you have one face, you want it to be the same thing across different video generations. Instead of trying to perfect that with your prompt, you can just generate an avatar and that way, everything will look very consistent. Here, we're seeing a bunch of examples of what other people made, as you can see, they're pretty high quality. I could go into each video, either remix it, meaning that combine my own take on this video or just download it or share it as it is. We have the option for both video and image. And with the image, you can actually see the full prompt. So basically the instruction that this person provided ElevenLabs in order to get this result. And the same thing goes for video as well. So here we have an animation. We have a long prompt, as well. Now the cool thing about ElevenLabs when it comes to image and video generation is that it combines all of your favorite AI models. So just to show you a preview, when you go down here, you may be familiar with most of these models. So we have GPT, we have craft, we have SDream, we have nano Banana, and so many others if you just scroll down. And that's what makes this platform a little bit more unique because apart from their own powerful models for music, voice and sound effect, it also combines all of the other models that you may have relied on and brought it onto one place. Apart from the models that generate content, we also have editing models. So upscaling, this is another famous tool. It's been brought into ElevenLabs. We also have background removal here as well, and there's some tabs for you where you could explore it. So if you want to upscale to edit, we can use these models, and they take credit as well, so you could go based on the cost here. Same thing applies to video that was images. We have video, and then we have lip sync. So lip sync is where you made the audio or you just uploaded it, and you wanted to basically generate a video where you match the audio. So if you made a voice over here or did a text-to-speech, you can now have that speech set by an avatar in the lip sync. So that's another cool feature. So let's dive into this new feature in the next couple of lessons, explore how they work, how you could get your first asset made, and how you could organize them moving forward. 32. Assets & Templates: So anything that you make is going to be saved in your assets folder. So if you click on that, we have different things already because we have been using ElevenLabs for a little bit. Any avatar that you make is going to be saved here. You flows for the AI agents are going to be saved here. Anything you upload to ElevenLabs will be saved here as well. So I have two images here. Can upload assets so that you could easily drag them into image or video generation. You can make a new folder where you put certain things inside, and this way, you can have a storage area online for your content. So I just made a folder. We could give it a different name. So I will say my video and images just so we could save the things that we're going to make in the next couple of lessons. This with other people, delete it if we need to. When we go in the folder, we get to upload things from our computer. And once we make something in image and video, we get to put it onto that area as well. So I could just go over here, download this, and then upload it to that folder to use as a reference for another project. If you're not sure how to create a full content using this new feature, ElevenLabs also has the templates tab, so that's right above the assets. And when you click on them, we have some easy drag and drop options for you to create different sorts of production. So a couple of examples here, we have let's look at the categories. Say we want to do mockups. I have this packing mockup generator. I could click on it, upload my blank product and the two D packaging design, and it's going to be turned into this really nicely polished up. So the templates essentially do something for you from start to finish. This is an example for a mockup generator, but we have a bunch of other things. So let's close this. Here we have a UGC video template. So if you want to create that sort of content, you can upload the inputs for this template. So the influencer portrait, which this is an example. But in the next couple of lessons, I'll show you how you could do the AI image generator in ElevenLabs, and that way you can use your own generations, too. Next, we have the product. So whatever product you want to show, you will put it right here. Choose your voice. This could be from your own voice library or the many voices available to you. And then basically the benefits that you want this model to now highlight. So here's an example. Then we have the shoot location. So right now, it's like a pink vanity corner, but you could put in the forest, an office or anything else. You're done, you can just click Generate, and you got yourself a nice UGC video. Now, of course, whatever you do, you still have the chance to edit it further, make refinements, and make it really your own. Down here, we can see how much credits this is going to take. But just to look at the flow, basically how this is going to generate that final product you, it's taking the following inputs that we looked at on the side. So if you're curious, you can always come into the flow section. So influence a portrait. This is an input, the product, the voice, perfect match, product benefits, and shoot location. So it's going to take these guys and write a script first, decide on the length and then have the models run. Then it creates a couple of versions. Down here, it's working with the voiceover and then it puts the two together for the final product. And then this is like the final final product. You could use the flow to figure out basically how it approaches this issue, it's kind of fun to look at too. So you can edit this flow as well, say you wanted to at some point, bring in another input for a second character. You can add that right here. We're going to learn about flows later on. But for now, this is something that you can explore. Now, these were the examples that we looked at and also things that other people have made. So you can see the image product. These are all really well done, really. You can just see the quality of the work. If you go to Generations, these are the things that you have made by filling these in and then hitting Generate. So right now there's nothing there. Think of templates as little tools that you could use on top of all the other ones, and they're really cool to work with. The only thing about them is that they do take a lot of credits to make sure that you're allocating enough credits for that and you are aware of how much is going to be used. So here's video effects for dubbing, upload your input video and then you take the dub out. Also make your own template if these tools are not really for your need, you can click on plus New Template. And here you can kind of talk with ElevenLabs to build your own flow. Now, again, I am going to dive into flows a little bit later. But basically, if you have something that you need to do repeatedly, like tasks that you have to do every day where you upload a video and it needs to be broken down into a vertical reel, horizontal reel, and maybe a square one, you can build a flow based on that and automate that process. So basically the input would be the the music and maybe some captions. You'd have 11 labs, break it down into three versions of the video, and then refine it with maybe more AI content, AI music, voice over, and deliver you that final product. That's an example of a flow. If you want, you could just leave it up to 11 labs completely. It's very user friendly. You don't need to use specific wording. You just say what you want, and then it will help you out along the way. So going back, we can now move on to how to start generating things with ElevenLabs. We're going to first get started with images. So right over here. And we're going to see how this whole interface works. So let's jump right in. 33. Image Generator: So the first option here is using this as a reference. So again, this is a template that I could give 11 labs to generating new. Maybe this time, I could have a penguin in this exact same pose. But in order to best describe the pose, I could just upload this image as reference for the tool to follow. That's a reference, and then we have upscale image that's going to increase the size and the resolution of the current image, make it high quality. For example, if it's a 500 by 500 image, I could upscale it to the size or even larger and use it for different purposes. You can recreate something, so basically put in this prompt, but see what comes out of it. So it's going to look a little bit different. We can use this for lips. So if I have an audio already, I could have this subject basically speak that audio. So it's going to this option is going to work on the lips, the movement, facial expressions, and all of that. And lastly, we have create video, so this could just be emotion. It could be the subject doing something or just standing there, things that we can define using this option. So a lot of cool things are being made here. We have mockups. We have art. We have some like photography techniques here. We have some designs, posters, illustrations, and so on forth. So we're going to do a very simple exercise right now where we try our first prompt and then see how we could build upon that. So what I'm going to do is type my subject. I want a woman, and I'm just going to send this in. So I'm trying to show you guys the power of a prompt and how well the output is going to be if you describe things better and you follow a certain format. I now need to see what this woman is doing. Is she sitting? Is she standing, lying down, what is happening? I need to mention that. So a woman sitting down on a chair. So this right here basically sets a scene. Not enough details, but we're just going to send this in with one of our models. So again, we can go to image, and there's a lot of models that you guys can explore. To get started, you can try one of the trending models. GPT right now is very famous for image. Nano Banana is also really good for characters, realism, and all of that. I'm going to stick with the newer model right here and click on that alone. So we have our model set. We can now describe basically set the aspect ratio for the final product. I'll go with the landscape shot. Next, we have the resolution, just to get a fast generation. I'm going to keep it to the minimum. Quality, I'll just go with high somewhere in the middle, and then we have generations. So how many versions do you want 11 labs to make based off of this prompt? I'll do four just so we can see how they differ. And lastly, we have this editing tool which helps improve your prompt. But since I'm trying to show you the difference between prompts, I'm going to turn that off. So just make sure you're image, put in your prompt, this exact prompt or something else that you prefer these settings, and let's send this in. Here are my different versions. We have a similar room. We have a similar subject. They're just in different shirts and different poses. But it looks like it's the same couch. There's a slight difference between them. So right now, it did give me a woman sitting on a chair, but it doesn't really have anything unique to it. It looks like a stock image, basically. But now, whatever I have in my mind, whatever I'm envisioning for this woman, I can now describe it building upon this prompt. So either click on this or type the prompt again. Now we're going to add a few more details. Since I can see the different versions, I could just cut this down to two, so it generates faster, and we can go through this a little quicker. So a woman sitting down on a chair, before we even mention the woman, we can see what she looks like. So in terms of age, in terms of maybe ethnicity, and then after that, what does her hair look like? What's the hair color, things that we can better describe the person. So we can see a young woman this case with, let's say, blonde hair this time, we'll do the opposite blonde hair and blue eyes sitting down on a chair. So right now, just the woman alone, we're adding more details that can make her basically pop from the following results, and we could even add some more. So she's sitting down on a chair, maybe, what is the chair? Exactly, what does it look like? Is there a certain color? We can mention that here. Sitting down on a velvety chair, and then we can mention more about her pose. You can say with a leaned forward pose and a confident face. Something like that. So more details, we can put that in. And we could also mention things about the background. So right now, I'm just going to send this in with the female only, not going to mention the background, and we can see how the new result is going to be a little bit different from the first one. This ended up being locked because probably because of the leaned forward pose, sometimes with wording, it may send some flares, but I just removed the lean forward thing and just did confident pose. So if you run into this problem, just alter your prompt a little bit. So this is the pose that I got. It's not exactly what I was looking for, but you can see the difference between this image and this image. The chair is indeed velvety. She is blonde, and she does have blue eyes. Now if I want, I could set the same character in this position, say, I really like this background. I could now use this pose as reference, say, this is what I was going for. Instead of trying to alter this, I could reference this image. So I'm just going to reference that first image and have my prompt as it is. And basically, the tool needs to replace her hair and the chair and give me a new photo. So there's my new model. She's exactly in this pose, but she's blond and the chair looks different. So this is what we have right now. It's looking pretty good. And since I have my model finalized, I can now use this as reference to have her do other things. For example, she has glasses on and she's on her phone. That could be a different pose in this exact setting. But by providing the reference, I could put less pressure on the prompt and more pressure on the visual. If you're not sure how to best write a prompt or you're struggling by struggling to refine it to what you need, you can turn on this feature. So basically, this is going to work through the current prompt, improve it, and give you something that is more suitable. So I could now click on Enhance, and it basically transformed my very simple prompt into something more detail. So flowing blond hair, using words that would better describe texture, light, contrast and all of that. Instead of just blue eyes, it's piercing blue eyes. So more detail. And then the color of the chair, what type of chair? And then the lighting, we didn't even mention the lighting, but it's doing that for us now. Something about the texture of the chair and, of course, the posture that she has. Different things you could do here as well, instead of just enhancing, you could further refine the pose. So a subtle, knowing smile gracing her lips. So this is more about her, the way she's sitting. And then, you know, you could go on and on by adding further details. So let's generate this with the same model and everything and see what we get. I did remove the reference image. Sometimes you may run into issues with certain models. For example, this is my prompt, but it's saying that I may have violated their terms of service. So certain models prevent you from using certain combination of words, and that's just to keep their community safe. But if I just switch to a different model, it may interpret my prompt in a different way that is not, you know, deemed harmful. So we could try maybe nano banana instead and try that out. So there we go. You can see that we didn't run into that problem. But here we can just see the big difference between the models. So this is the nano banana one. You know, it looks like she's in a studio, whereas the other one looked very natural. So changing models really helps when it comes to the visual delivery of that prompt. So that's one example. We could go down to Runway AI. That's another model and see what that looks like. We have Gen four Image. That's one of the runway AI models. And so far, we're just building upon our prompt. We first started with a woman sitting down on a chair. This is where we have ended up. And, you know, with different models switching between references, we are fine tuning that image that we want. So there is my new image. It's very different from the previous ones, but that's just how this model is interpreting my prompt. So a lot of models out there, each AI model is built differently, trained differently, so expect to see some big differences when you switch between the models. That's pretty much how the image generator works in ElevenLabs. The purpose of making one is so that you could maybe switch it into a video later on, combined with the audios that you have made already. It could be to create something with the templates that we looked at and you could either keep it within this platform, use it for different output or export it into something else externally. So maybe using it on another AI tool or just posting it somewhere. You could explore this tool. It's new, it's improving. But since there's a lot of external tools combined here, it really makes it convenient for you to make all of your content in one place. So that's the image generator. Let's move on to the video generator, which is the next option. 34. Video Generator: Similar to how you made images, you can now make videos. So that's the next tab right here. Once you switch over, there are a couple of different things, but the core concept is still the same in the sense that you still choose your model, the aspect ratio, the quality, and your references. The only difference between image and video is that you now have starting frames and frames and video and audio references. And of course, the duration because we're now working with motion. Way this works is that you can either just type what you want and have 11 labs generate a video for you purely based on the prompt, or you could have an image generated and then use that as your reference. So in this lesson, we're going to do a very simple exercise where we generate the image and then use that as our reference. We're just going to switch over here, and we're going to write something about our character. So I want a video where the subject is running away from the camera, and I want him to wear, like, a winter coat and maybe run in the snow. So that's my idea. I just have to bring it to life. And it all starts with what my character looks like. Switch over to image. I'm just going to type in a man wearing a winter coat looking very serious. And then I want this to be full body and hyper realistic. So there is my prompt. I'm going to switch to nano Banana, and then I'll keep this at 41k, and we'll do a vertical aspect ratio. So no references just typing this in, and we're going to generate this. So here are the characters. We have our first one, second, third and fourth. I could now choose the one I like and then use that as a reference to further define my character. So I'm just going to go with this guy. I think, you know, he has the full setup, and now I want him to hold a bag. So we're going to use this guy as our image reference, and then we're going to type in a new prompt. So a man holding a backpack in his hand, you can say which hand, what color the backpack is, but I'm going to keep it just like this. Same model, same settings. We're now going to send this in. Here are the models. They're now holding a backpack. Again, I didn't really say what color or which hand, but I think this one looks the best because it matches with his outfit and he has the gloves on. Now, this can be my image reference for the video, and I could also add in other references as we looked at it earlier. We can now switch over to video. And have this image as our starting frame. So I'm not going to use any reference for this example, but only the starting frame because that's technically a reference of its own. But actually, if we try to do both, you can see that we're not allowed to do that. So you cannot have a start end frame alongside image or video references. So now we just have to describe what this man is doing. So we could say man running away from camera chasing. I want him to be running really fast. Name settings. I'm just going to switch this to 10 seconds. And I'll keep this to 1720, and I'll use the S dance 2.5 model. Let's send this in. So that's going to take a while to generate. So at the same time, I'm going to have another example where we remove the start frame and use three references. So same prompt, we're just going to use our image reference, which is our character. Drop that in. Video reference, I could find a video online where I want my character to run that way. If we just go to the main page, explore video and search for Running Away, I could use something like this as my video reference. Like that. And for audio, I actually made one earlier where I wrote that man is running away, so running sound, heavy breathing, that sort of stuff. If you don't want to make your own audio, you could also look for something in the sound effects tab downloaded and simply upload it here. There's my audio. Again, same settings. I'm now just gonna hit Sent. So my videos are done. This is our first one, where we had him run away. So we can see that it provided the footstep noise, you know, suspenseful music, and, of course, him breathing a little bit at the start. Now, I could guide the audio, maybe have maybe describe the sound of the footsteps, which is going to be running in the snow. This here sounded like a rubber boots sound, which wasn't suitable. But you could see how it implemented that chase scene audio using just two words. So that's our first video. Let's take a look at our second video where we provided three types of reference. So that was our subject. You can see that he's wearing the same outfit, pretty much. Then we had our video reference, which is pretty identical. Lastly, we had our reference audio, which was just the heavy breathing. So all of that combined gave us this video. And you can see, at some point he turns back to the camera and he looks exactly like my subject. So that's the video generator that ElevenLabs recently introduced. It's really powerful. It's only getting better. You can try this with different models, but bear in mind the amount of credits that it's going to use because this is on the heavier side of production. You can see that this five second clip used 15,000 credits, and I did combined three references to make this video. Down here, you could see the size as well as the dimension, the resolution. So you could use this information to better plan your production without running out of credits every time. Now, let's go ahead and look at the lip sync feature inside ElevenLabs. 35. Lip Sync: We saw how to make video images, and now we're going to have our characters say something with lip sync. So that's the third tab, and all you do really is upload an avatar and a speech that you either made on this platform or had it from another place. You can guide the generation here, maybe talk about the way the facial expressions are made or the speed that the person is speaking. So we're going to just click on Avatar. You can either upload an image for your avatar or select one that already exists. If you want to make your own avatar, you can do that with the image generator tab. The one thing that you need to bear in mind is that you want the image to be close enough to the subject's face because a lot of the focus and generation is going to be around the face and the mouth. So this image right here, the character is a little further away from the camera, so we may end up with some glitching the face when ElevenLabs is trying to make the characters speak. But when you go to the homepage, you can see that we have some avatars. So when we click on View All, notice how close the subject is to the camera, and this is going to really help with the most basically making natural gestures, natural lip movement, and all of that. Basically, you're going to click on Avatar, select Avatar, and then choose whichever you want. Once you click on them, you actually get more options about the positioning of the subject, the clothing they're wearing, the location. And you can see that I have a lot of options. So if I go with this one, I could click on you style and then go for the speech. It could be text-to-speech or something that I have uploaded. So if you do text-to-speech, this is our saved paragraph from a couple of lessons ago, and I could just either use the speech that I made in that lesson or generate a new one with a different voice. I'm just going to click on the latest one, and you could just click on text-to-speech or upload your audio. If you want to make your own Avatar, you just have to click on New and either create the face from a prompt or upload reference images. So basically, if I go into any of these, you can see that there is a lot of different angles from that one character. And that's basically what you want with your avatar. So if I head over to upload images, I'm going to need different angles of my subject. And that's something that again, I could make in ElevenLabs or something I could do with my own camera. So upload the images. Ten is recommended. You can use these as reference if you want to take your own images, or you could let 11 labs create each one for you and just upload them here. You can give the Avatar a name and a voice, and that's it. You just click on Create Avatar. If you want to start from a prompt, you don't need to upload any images or any references at all. You just describe what they look like, the way they move, and that is. Only issue with this is that you may have to do a lot of editing to have 11 labs understand your ideal avatar. So if you have the images or if you've made them here, it's better to go for this option. So I just chose my avatar and I'm going to do a text-to-speech, something very simple. You just have to change the voice to, you know, the avatar's voice because this is a template. So I'm just going to cut the majority of this paragraph and have him say this very informative sentence, generate speech. Last Friday, students from Sunshine Elementary School planted 20 young trees in the school garden to keep the environment clean, green, and cool. Once I'm happy with this, I could use the speech and then send that in. Down here, we do have a few options. We already know what these two are, but this guy is basically the model for lip syncing. So it's the same idea as the image and the video models. Only these are made for a different purpose. You can go with any of these, but I just went with the most trending one. Now, while this is loading, we could spend some time generating our avatar to see how well that works. Our avatar is now saying our speech. Last Friday, students from Sunshine Elementary School planted 20 young trees in the school garden to keep the environment clean, green, and cool. You could see the lip movement. According to the words that are being said, he does have some motion in his body. He looks at the camera, and it has that podcast like wipe to it. Last Friday, students from Sunshine Elementary School so, of course, the motion is not that natural. This is AI lip syncing, but it could be helpful for certain types of content. And you can explore this with other avatars, even make your own avatar. You can see how you can combine your audio with either an avatar like this or a video content that we made in the previous lesson. So these are two options to make your clips more engaging while showcasing the quality of your AI-generated voices. Mm hm 36. Upscaling with ElevenLabs: If you ever ended up with an image that is not the best quality or you just want to change the size of it, you can now use upscaling in ElevenLabs. So I just went to my assets folder and I uploaded an image that I made with AI. So currently, it looks like this. It's a closeup of a man's eye, but I want to add more detail. I want to make it larger. And instead of going to another tool paying for it, I could just, you know, come in here with the same subscription and upscale. Click on Upscale, and it's going to bring you to the image and video tab. Right now, this is the image that we're going to upscale, and it just automatically chose this model. So this is the only model right now that does the upscaling. And then on the right side, we can decide the size of it. So I'm just going to go all the way to the max just so we can see the difference, and that's pretty much it. So you don't need to type anything or give any other references. You just send it in, and it's going to upscale your image. It's done upscaling. It's up by four times. This is the original image that I made. It looks like this, pretty detailed already. But pay attention to the size here. This is the size and the current dimension in pixels of this image. I then upscaled it using the model, and now we can look at the size as well as the dimension again in pixels. So that is an improvement by four times, and I can now use this for bigger projects without losing any quality. That's pretty much how you could upscale. You could further upscale upscaled image. So I'm just going to go to the one that I just upscale. I still have the option to make it bigger. So I could go to this size now and send that in. But that's only going to build upon the previously upscaled image. It's a pretty straightforward tool if you felt like, you know, the image that you cut is just lacking detail. It could be more crisp. You can use that feature to, you know, make it highly detailed, much larger, but bear in mind that the size is also going to be increased. So you can go ahead and try that out with the images that we've been making and compare the different sizes. 37. Flows: All of the new tools here really give you the possibility to expand your creative outputs. So this could be music, sound effects, video, image, and the classic speech-to-text or text-to-speech tools. But flows takes all of that to the next level and allows you to basically have a canvas for generating the content with all of these tools. So it's a very different sort of layout to what we've been used to this whole time. We're going to be exploring it in this lesson. We know that studio can be used for larger productions, but flows gives you a whole new layout where you get to drag and drop different tools, connect them, and kind of work with them as if they're nodes. So if you're familiar with nodes, it's pretty much that same idea. You create one, you connect it, and you build a flow. So here are some examples you can think of a final product, but then go backwards and see what the different components were. So this guy right here, if we click on it, we can see the flows that is behind this production. It looks like a lot. And these are things that we may not pay attention to because we're only seeing the final product. So basically, if we zoom in, you can see how everything came about and where the starting points were. So we have the initial designs. We have the text. You can use the space bar to move the canvas. We then have different text, again, some more text because this is an explainer video. Um, and then I'm going to find the visuals. So these are the visuals which were not generated, and then I could move each of these notes. So right now if you look at this, it's a lot, but I'm going to show you how you could make your own flow, and then this whole thing will make sense. So this is just an example of what it looks like. If we go back and hover over this, this is what that flow chart two. So let's click on New flow, and we're going to be met with a blank Canvas. So just to go over the interface real quick, we have the menu on the left side. So you can make a new flow, quick actions, give it a new name, duplicate share, create a template based on this flow that we're going to make. Even see your version history. So you could go back to an older version. First, let's give our flow a name. We're going to call this my flower video, maybe. And the concept that I have in my mind is like a flower pot where a new plant is growing within. So that is my concept, and I'm going to make a flow based on that. So before you click or write anything, you're going to need some different components that are going to lead to that final product. I know that my final product is going to have a visual, a video, and a video is going to need some image references. So that's already two things. It's going to need an audio. Maybe that could be a voiceover, as well as some background music, and I want to add in some sound effects as well. To help write the voiceover content, I'm also going to use AI for that. So if you count all of those, they're going to be multiple components. But from now on, I'll call those components nodes. So simply right click on this Canvas, you're going to have an option to add a new node. You can also hit N on your keyboard for a quick action. So notes can differ. It could be an image, video, text-to-speech, lip sync, and other things. So these are all things that we have looked at so far. It's nothing new. The only difference is that there are now smaller parts that play a role in a bigger production. So these are all like actors in a bigger movie, basically. So like I said, I'm going to need an image to get started. So let's click on this, and we're going to get this box. Now, the interface here is what we've been doing this whole time. You choose your model, the ratio, resolution, quality, and then you write your prompt, you hit Run. Only difference here is that we have these things on the side. So these are the connectors. So what is going to be done with this image after it's done? Is it going to link to a video production or is it going to be the result of a text box? So this is where the node idea is going to make more sense. So instead of writing something here, I'm going to make a text no that way I could have different models, make that same prompt, and then I could choose the one that is best for me. That's my image right here. Let's right click again, add node, and we're going to go down to or just go here for text and choose text. That's my text box. I could start writing or I could use the help of a large language model. So say I don't know how to write my prompt for that flower pot, I could create another node. I need the help of a LLM. Let's put that here. Let's go down. And now I have three things. And just to tell you what the flow is like so far, we're going to have a prompt generated by Gemini or any other model that we want. That's going to lead to this text box where I could see my finalized prompt. This text prompt is going to lead to the image generation. So the image is going to be made based on this guy, and this is going to be the result of this guy. So there is a connector, 0.10 0.2 0.3. So let's start right here with our LLM. You can see we still have some reference points, but I'm just going to have this be the starting point. So let's say give me an image prompt for a nice plant pot that has no plant in it and is only soil. Make this indoors with good lighting. So that's my prompt. I just basically have an idea so far. I'm not going to have the final prompt here, and I'm just going to choose my model now. So this could be anything. Let's say you want to do Gemini is fine here. Then you can choose the mode for that model. So if you're each of these models are going to have their own mode. So, for example, if I chose Claude, I'm going to have different options for the reasoning and thinking, and these will all result in different things because we're dealing with different models. And then we have more here as well. So let's just go with the Gemini 3.5. I'll do minimal. There's not a lot of thinking needed. Here we can decide on the length. This could be automated. I'm going to leave it entirely up to Gemini to decide the length of my final prompt, and then we can start running this. But before I do that, I'm going to point the output, which is on the right side to my text box. So these are inputs. These are outputs. Use the space bar to move the canvas around. So I could also use this actually as my input idea. So let's break this down even more. Let's copy what I wrote and put it here instead and use this as an input. So just connect it to. That's our first connection. You can see that the outline became darker and we're seeing an actual line. I could always cut this connection so they're not related anymore, or I could just make a new one like that. So now it's doing at text. That's the name of this box. If I change the name, let's call this my idea. That's going to be changed here as well. So it's telling this box to look at this guy to get started with that prompt. So at my idea, we're going to hit Run now. Using our Gemini 3.5. So it's going to generate a nice prompt for us that now I could connect to my image node. So I'll call this. So this name is fine. I'm now going to output this as a text input for my image node. If you want, you could also make your changes, run it again and then do this, but I'm going to be using this prompt as it is. So now notice it says at LLM right here, and I'm going to have this image node generate with my favorite model. So I'll do nano Banana, choose the Aspec ratio. I'll do a square for now, the quality and everything, things that we have used so far, and I will hit Run. So that's going to do its thing. But say I want to do multiple models. So this is not enough for me. I also want to try JAGBT. I want to try basically a bunch of other models to determine which one is best for my project. We are going to right click again or just click and drag. And once you let go, it's going to ask you, well, what is this going to lead to? So I'll just do image generation again. And this time we can choose something else. So let's go for Cream and then hit Run. Now my text is leading to two things. We had our first image, second image. I could do a third one, a fourth one. As long as they are connected to this, this is going to be their prompt. You could, of course, write the prompt directly here, but creating a flow like this gives you a lot more flexibility when it comes to editing and, you know, refining your project. So we're going to have this guy do its thing. You can see that it's about 30% done right now, and I could continue building based off of this. Now, you could also add a image reference by right clicking, adding another image. So let's go to image right here. Click Image generation. Now, over here, I could just write something. We can say basically this idea, but not counting the pot. I want to talk about the room. So I'm gonna say a cozy living room with open windows and blue furniture, something that could be a little different from what we have. So you could again use a large language model to help you write the prompt, but I'm going to do something very simple, and I will use nano Banana for this run it. So here's our second image. You can see how different they are. And I could now connect this to something else. But first, I want to use this room and have it be an image reference for this guy. Say, This is more ideal for me rather than this. I could now connect this image output as an image input for this pot. So let's move this here. Let's give it a name, actually. So I'll say this is image reference. This is Image one. Let's say. That's image two, just so you know what I'm referring to. All right. Once we have our image labeled, we can now take this as an input for a video. So we learned about videos. We learned that we have starting frames and image references involved. So this could now lead to my video footage. So click and drag, let go, and we're going to choose Video generation. There it is. Here you can see that we have multiple inputs, just like we saw in the previous lessons, we have starting frame, end frame, image reference, video reference, and audio reference. We also have the prompt input. This could be whatever you type here, or it could be a text box like this. I'm just going to have this be a image reference. So let's hit X and then connect it again using this. So that's what it's going to be. This way, I could reconnect it to whichever input I want. But I actually realized that we wanted to redo the background. So now that I have two things connected. Original prompt and this new background, we're going to hit run again. So that's going to redo the whole work and give me the same flower pot. So there's my flower pot. We could see that it integrated the blue, not so much the couch, but that's okay for me. Since I just wanted to show you guys how an input reference is going to change the result. So now that I have a blue flower pot, I'm going to use it as an image reference or a starting frame. Ing on what you need. And I'm going to have one of these models. I'll do seed dance. Have a basically seed sprout into a flour from this very pot. So seed sprouts in the pot turning into a nice red. Flower. Choose the settings. I'll do squares, since that's what we're doing so far, the resolution, and then you can choose the duration. I'm going to mute it here because I want to make another node for the audio. So let's run this. If you want, you could have another large language model. Node, be here to create a better prompt for you. But we're just going to do what we have right now and see what comes as we saw, we could also make multiple notes for video where we try different models, and then we can choose the one that we prefer. Now, while this is running, we are going to work on the audio. So somewhere along the canvas, I'm going to write click. Add node, but this time, we're going to go for audio. This could be text-to-speech, sound effect, music, or anything else that you want to get started. We're going to right click here and make a simple audio note. I'll do text-to-speech, and we'll do a very simple sentence. So we can say new beginnings and new flowers all here to stay. Something like that, choose your voice, and you can also work with the stability, choose your model, and then click on Run. So it's 3 seconds. New beginnings and new flowers all here to stay. Very good. And I could then combine this into, you know, my video right here and get my final result. So just drag this audio output and put it in as an audio input. Move this to the side. And so far, we have this, you know, mixed up map all leading to our final exported footage. Bear in mind that each of these do take their own credits. So make sure that you are allocated enough credits for that. I'm going to leave mine like this, but if you hit rerun, it's going to apply this audio into the footage as well. Once you're done with your video, although this is not really finished yet, you have the option to copy it, duplicate it, maybe connect it to another where you have someone grab the pot or a different scene. You can make a template out of this whole thing with this button. You can download it when it's done. And since this is a canvas, you can also invite other team members to come in here and, you know, help you along with this project. There are some tools for that. We have our Move tool, our select tool. We have our Move tool, which is what we've been using with a Space Bar. Have the comment tool. Maybe a friend comes in here and they think that the couch should be read. They could leave a comment, and it will go here. So make red, save the comment, and now it goes here. I could resolve the comment, disagree or agree, or leave another comment. You can also delete people's comment if you want. Now if you're not sure how to get started with all these notes, you can also just start chatting in this box. So say you can write about the final scene that you want, a flower pot where a new plant is growing. You can write it here, and then ElevenLabs will guide that production for you. If you have references saved on your local device, you can attach them here, and you could send that request in. Now, say this is a workflow that I do very often for other projects for my social media maybe. Instead of creating this whole node layout, I could create a template. I click on this, I just have to identify the inputs, outputs, and the publish format. So things that only have something coming out the right side are the inputs. So this guy had no inputs, therefore, it is its own input. So it's just a lone like this. This is my input. I can select it, and it's going to be marked as orange. And then I could do the same thing with output, give it a name, and then I could publish my template. So that's something you could do as well. Here's my video. We have, you know, a seed sprouting. It looks kind of magical, but it is a red flower. We could then extend this to a different node. So a video extender note, if we want to turn it from 4 seconds to 20 seconds, we could do dubbing, mix it with an audio, use it as a reference. And so much more can be done. So go ahead and try out the flows. It is a new tool, and it's really cool, actually. And you could really work on larger production, combine different tools as notes, and build a final product where you use different things in ElevenLabs to tell a story, to promote a product, do a podcast, or anything else that you want to do. 38. What is ElevenAgents: So we looked at the creative side of ElevenLabs. We looked at voice. We looked at image, we looked at video, and so many other tools on the left side. But they also introduce something called ElevenAgents, and that's something you can switch to by going down here. So just click on that and we're going to go over to the next interface, basically. So ElevenAgents are basically AI agents that you can use to automate a certain task, either integrating the tools that ElevenLabs has or, you know, just keeping it as a simple agent. So one of the templates that you can look at is customer support. That's just an example of an 11 agent. And on the side, it actually tells you so much about the workflow behind it, how it's working, and you can also preview it here. So right now, I'm going to pretend that, you know, I need support, and this is currently taking in two integrations. So Saleforce and Zendas we have many different options here. But the main feature is basically the tools that you already know how to use being integrated in these workflows. So once we see an example, it's going to make more sense. Hey, this is Jamie from support. What can I help you with today? I need help with a technical issue. I'd be happy to help you with that technical issue. Could you tell me a bit more about what you're experiencing and maybe share your name so I can look into your account? My name is Alice and my computer broke. I'm sorry to hear your computer broke, Alice, but since I support our AI platforms APIs and integrations, I can only assist with issues like API errors, integration problems or performance concerns. So that's just an example. You can see, even though I didn't really ask what the support is for, it still understood my concern, and it took in all those inputs. So the start was me pressing on the call button. The first thing was to identify the issue. It opened up warmly, which it did. And then I just said that I had a technical issue. Then we went here to propose one concrete first step, and then you could either resolve or escalate. Say, it says, Okay, you need to provide us with a serial number, and then I agree to them. They're going to resolve this now, and then it ends. If I said that I have an account problem and not so much a technical issue, it's going to take a different route. So it's going to look at billing with my name and then see if it reached a solution, either go with the solution it has or look further, and then end the call. So this whole map is basically how this AI agent is going about being a customer support. We have different things, too, something a little bit less technical is a language practice tutor. So you can start again, and the first thing it's going to do is access your level for that language. If I'm beginner, it goes this way. If I'm advanced, it goes this way. Beginners, it's going to have a certain mode. So maybe using certain dialects, steering away from large vocabularies, speaking, slower, those things. And then the session ends. It gives us a summary. So lots of templates you could look at, but, of course, we could always make our own by creating an agent. So you can browse the templates, make adjustment or create one from scratch. So we're going to learn all about that. But just to give you a little bit of a tour, we have the different categories on the left side to help us get started. We have creating agents, which is where we were configuration, the knowledge bases that it has, I could create and upload some documents here that would be used for the agents. For example, if I have a agent that is supposed to help customers with broken laptops, that agent needs to know a lot about laptops. So instead of typing everything in, I could just have a document uploaded here and then connect this base to the agent. So it can go back to this, learn a little bit more about laptops and then get back to that collar. Could be a URL, could you trust a website. You can put it here. You can upload a file, type something in, create a folder with multiple files or sync documents from other platforms. We have 21 megabytes of storage here, which is perfectly it's plenty for documents, and you could sort and search as well. Of our different tools here that you could add in. So any sort of integration that you want to do. So I would have to add a connection first. We have a lot of things that currently work with ElevenLabs, and all you have to do is sign in, make that connection, and it will be available for you to use in your agent building process. So these are the different categories, but we have all of them listed here. You can search, and once you again sign in, they're going to be listed here and you just select them. So that's an integration tool. We have client tools. Client tools are things that require an input. So, for example, the template that we looked at where the agent was asking about the type of issue that I was having, that is a client tool. So it's like technical issue or just type of issue, wait for user response, and then do certain things. We can add parameters, different variables. Even edit this as a JSOd file. We have webhook as well and then a list of other integrations that you could add. Of course, we have our voices. This is 11 labs after all. These can be integrated as the voices of your agents. To monitor, you can look at the conversations that are happening as you deploy agents, the users, and then the different tests that you do. So when you make an agent, you do want to test it out, see if it works fine, and those will all be logged here. Deploy, you are going to have the following. So it could be deployed with phone numbers to Whatsapps or even batch calling. So we have all the levels that we need to deploy a successful AI agent. Now, we're first going to start with our agent. We're going to build one in the next lesson, integrate a few tools, test it out, and then deploy it. So you guys can follow along with this simple agent that we're going to build and that we are going to have a better understanding of how this works and how you can apply it to your daily work. O 39. Create First Agent: So let's build our first agent. This is going to be a very simple agent, just to show you how it works, and then based off of that, you could build more complex agents and integrate different things. So to create an agent, you click on this button right here. And instead of choosing templates, we're going to create a blank one. Now, the templates, we already looked at them, you can choose one that is similar to what you need and simply switch out the things that are irrelevant. We're going to start with a blank agent. So first, let's give our agent a name. Going to say information provider, and we'll call him Ben. So you can choose here for this to be a chat only agent or an audio agent. So by default, it is audio because this is ElevenLabs, but if you want to exclude that, you just keep it at chat only. So basically, it would be similar to the chatbots that appear on the side of the website. But the audio is going to be something like a call in a way, where it will wait for a user input speech in order to operate. I will leave this off, and we're going to create Ben. So here's the new agent. We can see that there's a bunch of things around that we can now change. So on the right is the preview of our agent. And because we said we want audio, you can see that it has, like, a little phone call symbol, and I could talk with Ben right now and have him help me with something. Now we do need to set Ben up because right now he's just an avoid, basically. The first thing we need to do is tell Ben what they are. So this could be you are a customer support agent. You are a personal assistant or something like that. Whatever you want to do, you can put it right here. We also have our generated with AI option if you're not sure how to go about writing the system prompt. But basically, if you click on the little book, it's going to show you the official guide. So you do need to set up the environment, the personality, and different aspects of this agent. So you can see tone is one of them. The goal is one of them, and the way that this agent should interact and process the inputs. So you could read the guide if you want to learn more. But to get started, I'm just going to have Ben act like an automated voice agent whenever someone calls the business, and our imaginary business is a hair salon. So I did make a knowledge based document, and that is available for you guys to download. And I'll show you where you can upload that. But for now, we are going to tell Ben Act. So let's say you're a helpful assistant for a hair salon and people will call to ask about business operations and hours. So very simple, his job isn't that hard. I could now use the tool here to enhance this further. So generate the AI. Let's actually copy this and put him in and see what prompt ElevenLabs will give us that is more suitable. So you can see that it followed the same guide have personality, the environment. So who are they interacting with, the tone of voice here, the goal, the whole purpose of this agent, and guard rails. So the only thing that I need to change is, you know, the name. So you are Ben. And I forget what the name of our hair salon is called. Here's the document that I was telling you guys about. This is what I ended up calling the hair salon, so I'm just going to copy that in. The Locke hair salon or Hair Studio, cut that out. And let's check the rest. He knows everything, and then you're interacting with colors, and I'm just going to replace the name as much as I can. So that's the name of our hair salon. I'll do Control F, look for the old name, and just switch it out. This is Ben, not Stella. So just do a little cleaning up. Alright, so no more glamour. And I'm just gonna check for Stella in case there's more, and there is not. Alright, so we have everything set up for us. I didn't have to do much, as you saw. Then we're going to decide on the first message. So when the person calls Ben, what should be the response? So, hello, how can I help you today? I could say hello, I'm Ben. How can I help you today? Now, this option allows the user to kind of cut him off and, you know, say their urgent message. But if you want him to go on and say what he wants, you can just turn this off. So this is all good. Next is the voice. So we have a lot of voices to choose from. I'm going to go with Eric because it's one of the defaults, and it's pretty smooth for this purpose. We do have something called expressive mode, which is going to be more, as they call it, emotionally intelligent. So when it comes to pauses to, like, making different sounds, it's going to be less robotic. So if the user said, I fell down the stairs, they're not going to be like, Oh, that's great. Let's do this. They're going to be more expressive and sound more human. So that's something you could add if you want, or you could completely dismiss it. Up to you. So I will enable it for my purpose, and I'll just close this right now. You can see it's turned on, and it's Eric. Next, we have the language. We can only have people speak in English with Ben or add other languages. That's up to you. And then we have our language model. So there are a couple of things to get started, I'm going to keep the default. But if you want to build your own model, oh, that's something that you can do right here. It is a little bit more intensive, so I'll leave this out for now just so we can get our simple agent going. So that's the default and then agent behavior. You can basically set up behavior where it works with another tool, and there is a routine that this agent needs to follow. For example, we all know WhatsApp. That's like a texting platform. So if someone texts them, they have to first read the text and then reply to the text and also call. So these are things that WhatsApp allows people to do. And therefore, that agent needs to follow that behavior. This is going to be different for fresh desk, so that's going to be responding to tickets only. So each of these guys are a little different. And if you choose them, there's going to be additional options for them. But again, I'm going to keep everything as default just so we can get started. So turn that off, and we have now made our agent. Let's go ahead and add a knowledge base, which is where we create a space full of information about our brand. So in my case, that's the hair studio, and I need to provide Ben with business hours, the type of services so that when someone calls our number and he picks up, he knows exactly how to respond to the questions. Mmm. 40. Integrate: So, we made our agent in the previous lesson. Let's go ahead and give Ben some knowledge. We're just going to head down to knowledge base, and we're going to upload the document that I mentioned that is available for you guys. This could also be a URL or a different file. Mine is going to be a file, so let's just upload it now. So there's the file. We can keep it to our current folder. We're just going to add the file. So right now, whatever you add here is going to the knowledge base for Ben because that's the workspace that we are in. So if you want to remove something, you could just click detach from agent, and that's going to be out of reach for Ben. But I do want to keep that, so we're just going to now use this integration to help Ben with his job, basically. Apart from document as a base, you can integrate different tools for your agents. So there's a bunch of things that you could connect. I'm just going to go over to tools to add a tool, you are going to need something called API keys, which is the unique key that you get, and you bring into this platform to help the connection between ElevenLabs and whatever tool that you're trying to connect. So, for example, if I go to Ad Tool Integrations add a new connection, let's find something fairly easy. I'll go with Exa. I could give this a name, web search. I need an API key. So let's head over to EA and get that for ourselves. You just search a API Key, you're going to be directed to this link. Create APIKey and we're going to just create an account. This will be the case for any tool that you choose. So that's what I'm going to do real quick and then we're going to get a key for Bed. So I just signed in using an email, and we're going to create our account. I'll just skip this for now. And I'm just going to go down to APIKeys and create one. So I'll call this ElevenLabs. Create the key. Make sure you copy it, go back here, and we're just going to paste it in this area. Once you're done, you can click on Connect, and now it's all good to go. So once we have this connection, we can decide on the tool and the service that we want EA to give us. And for my case, that is a search. So click on search and click Create. So once we're done in tools, we could just click on EA and basically say how Ben should use this tool. So the condition prompt is the following. This is the default. So we can say choose content type, use highlights, recommended. I think this is fine for now because the user could ask anything from Ben, and I don't want to restrict it here. So I'm just going to let this be as it is. And we have now made our agent an integrated a tool. In the next lesson, we're going to try Ben out and see what sort of changes we can make and then how we can publish it onto somewhere else where other people can use it. 41. Testing: M So Ben is all set up and we now have to test to see how well he performs. So in the same page for agents, you can basically start talking to Ben. You can see on the right side, we have the little basically caller interface where I get to just speak into the microphone and see how Ben responds to me. So let's start a call. Hello, I'm Ben. How can I help you today? Hello. When is the business open? We are closed on Mondays, but open from 9:00 A.M. To 6:00 P.M. On Tuesdays and Wednesdays, 9:00 A.M. To 8:00 P.M. On Thursdays and Fridays. 8:00 A.M. To 5:00 P.M. On Saturdays and 10:00 A.M. To 3:00 P.M. On Sundays. Does that help you plan your visit, or is there anything else I can assist you with? Yes. Can you tell me a little bit about the latest hair trends that I could tell my stylist at the business? I don't have specific information on the latest hair trends in my current resources, but our stylists are very knowledgeable and can definitely discuss the newest styles with you when you arrive. Would you like to go ahead and book an appointment so you can chat with him about what's popular right now? No. Can you do a web search about it? I'm unable to perform web searches, but I'd be happy to connect you with one of our stylists who can share the latest trends. Would you like me to transfer you to someone who can help with that? So I just ended the call, but so far, when it came to talking about the business, Ben was able to help us out. But when it came to the web search, he still didn't have the permission to do so. So what we're going to do is go back to the system prompt and include the web search aspect. So I'm just gonna go here and find out about the goal. So welcome. It's fine. Okay, so we have appointments, location, directions, general inquiries, and then we could add another one. I'll just copy the same line just so that the format is the same. We're going to say for web search inquiries, answer questions using A for websearch and provide user with the highlights. So let's try this again and see what happens. I'm gonna close this right now and then start another conversation. Let's actually also change his first message just so we can see the update. So how can I help you today? Do you want to style your hair? Alright? Let's try this out. Hello, I'm Ben. How can I help you today? Do you want to style your hair? Yes, tell me more about the business services and when I could come in to get my hair done. We offer a wonderful range of services, including women's, men's and kids haircuts blow I don't want any kids, just for adult female. Oh, absolutely. We have fantastic services specifically for adult women, such as our signature women's haircut, blowouts and various coloring options like Baliage or root touch ups. You can visit us Tuesday through Wednesday from 9:00 A.M. To 6:00 P.M. Thursday and Friday until 8:00 P.M. Saturday from 8:00 A.M. To 5:00 P.M. Or Sunday from 10:00 A.M. To 3:00 P.M. As we are closed on Mondays. Does that sound like a good time for you to come in? Yes, but do a web search about the latest hair trends so I could come in and tell my stylists about it. That is a fantastic idea. The latest trends for late 2026 are all about polished, healthy looking hair with a natural feel. Popular styles include sleek blowouts, modern chignons and the boyfriend Bob, while colors are shifting toward rich, glossy brunettes, warm cognac brond and soft coppers. Would you like to book an appointment to discuss trying one of these looks with one of our stylists? Yes, book me an appointment. I would be happy to help you get that scheduled. To find the perfect time, could you please tell me your name, which service you are interested in and what day or time works best for you. Also, do you have a preference for a specific stylist or would you like to see whoever is available? Actually, I changed my mind and I don't want an appointment. No problem at all. I completely understand that plans can change. If you ever decide you'd like to book an appointment later or have any other questions about our services, please feel free to call us back. Is there anything else I can help you with today? No, that is all. You are very welcome. Have a wonderful day, and we look forward to seeing you at Lux Hair Studio whenever you are ready. Goodbye. So as you can see, that was a very natural way of speaking for Ben, and now it was able to look onto the web via the tool that we connected and provide me some information. And I did interrupt Ben at the start, and that is because we have this turned on. Again, you can turn it off if you want your agent to just say their sentences fully. And the back and forth between me and the agent also shows the power of 11 agents. You can see that I went to make an appointment, but then I changed my mind, and been was able to use the correct tags to respond to me as a user. So understanding, friendly, and here for goodbye, went for warm. And then when I finally said yes to the appointment, it was cheerful. So these are all signs that this conversation went very well, and we set up Ben correctly. So we tested Ben, and we made some adjustments. Also do more intensive tests. So that is where you set up a template, and you have that go on for repeat until a problem is found. For example, I could set up a test where Ben is called five times in a minute, and I could see how well he jumps back and forth with the five callers. But that is something a little bit more intensive, and for now I want to shift the focus towards publishing this new agent. 42. Publishing : M So Ben is all set up and we now have to test to see how well he performs. So in the same page for agents, you can basically start talking to Ben. You can see on the right side, we have the little basically caller interface where I get to just speak into the microphone and see how Ben responds to me. So let's start a call. Hello, I'm Ben. How can I help you today? Hello. When is the business open? We are closed on Mondays, but open from 9:00 A.M. To 6:00 P.M. On Tuesdays and Wednesdays, 9:00 A.M. To 8:00 P.M. On Thursdays and Fridays. 8:00 A.M. To 5:00 P.M. On Saturdays and 10:00 A.M. To 3:00 P.M. On Sundays. Does that help you plan your visit, or is there anything else I can assist you with? Yes. Can you tell me a little bit about the latest hair trends that I could tell my stylist at the business? I don't have specific information on the latest hair trends in my current resources, but our stylists are very knowledgeable and can definitely discuss the newest styles with you when you arrive. Would you like to go ahead and book an appointment so you can chat with him about what's popular right now? No. Can you do a web search about it? I'm unable to perform web searches, but I'd be happy to connect you with one of our stylists who can share the latest trends. Would you like me to transfer you to someone who can help with that? So I just ended the call, but so far, when it came to talking about the business, Ben was able to help us out. But when it came to the web search, he still didn't have the permission to do so. So what we're going to do is go back to the system prompt and include the web search aspect. So I'm just going to go here and find out about the goal. So welcome. It's fine. Okay, so we have appointments, location, directions, general inquiries. And then we could add another one. I'll just copy the same line just so that the format is the same. We're going to say for web search inquiries, answer questions using A for web search and provide user with the highlights. So let's try this again and see what happens. I'm gonna close this right now and then start another conversation. Let's actually also change his first message just so we can see the update. So how can I help you today? Do you want to style your hair? Alright? Let's try this out. Hello, I'm Ben. How can I help you today? Do you want to style your hair? Yes, tell me more about the business services and when I could come in to get my hair done. We offer a wonderful range of services, including women's, men's and kids haircuts blow I don't want any kids, just for adult female. Oh, absolutely. We have fantastic services specifically for adult women, such as our signature women's haircut, blowouts and various coloring options like Baliage or root touch ups. You can visit us Tuesday through Wednesday from 9:00 A.M. To 6:00 P.M. Thursday and Friday until 8:00 P.M. Saturday from 8:00 A.M. To 5:00 P.M. Or Sunday from 10:00 A.M. To 3:00 P.M. As we are closed on Mondays. Does that sound like a good time for you to come in? Yes, but do a web search about the latest hair trends so I could come in and tell my stylists about it. That is a fantastic idea. The latest trends for late 2026 are all about polished, healthy looking hair with a natural feel. Popular styles include sleek blowouts, modern chignons and the boyfriend Bob, while colors are shifting toward rich, glossy brunettes, warm cognac brond and soft coppers. Would you like to book an appointment to discuss trying one of these looks with one of our stylists? Yes, book me an appointment. I would be happy to help you get that scheduled. To find the perfect time, could you please tell me your name, which service you are interested in and what day or time works best for you. Also, do you have a preference for a specific stylist or would you like to see whoever is available? Actually, I changed my mind and I don't want an appointment. No problem at all. I completely understand that plans can change. If you ever decide you'd like to book an appointment later or have any other questions about our services, please feel free to call us back. Is there anything else I can help you with today? No, that is all. You are very welcome. Have a wonderful day, and we look forward to seeing you at Lux Hair Studio whenever you are ready. Goodbye. So as you can see, that was a very natural way of speaking for Ben, and now he was able to look onto the web via the tool that we connected and provide me some information. And I did interrupt Ben at the start, and that is because we have this turned on. Again, you can turn it off if you want your agent to just say their sentences fully. And the back and forth between me and the agent also shows the power of 11 agents. You can see that I went to make an appointment, but then I changed my mind, and been was able to use the correct tags to respond to me as a user. So understanding, friendly, and here for goodbye, went for warm. And then when I finally said yes to the appointment, it was cheerful. So these are all signs that this conversation went very well, and we set up Ben correctly. So we tested Ben, and we made some adjustments. Can also do more intensive tests. So that is where you set up a template and you have that go on for repeat until a problem is found. For example, I could set up a test where Ben is called five times in a minute, and I could see how well he jumps back and forth with the five callers. But that is something a little bit more intensive, and for now, I want to shift the focus towards publishing this new agent. 43. Other AI Tools to Use with ElevenLabs
: Now that we know how to fully use ElevenLabs to make different sorts of audios, ranging from a voiceover to an audio book, we're now ready to create audios in the best way possible. But you can also take your skills further by combining different AI tools with ElevenLabs. So ElevenLabs is mainly known for audio. As we saw, it's really good and it's constantly improving. But how can you combine the other very good AI tools to make some amazing projects? So I'm going to recommend some tools that you may or may not know already and tell you how you could combine it with ElevenLabs. The first tool that we already saw how to use was Chat GPT. I'm sure you're mostly familiar with this. But what you can do with it is first of all, turn A URL into a summarized version so that you could use it for a podcast or an audio book. As we also saw, you're able to change a language and to add in few elements that can enhance the speech that 11 Lab creates. Which at GPT, you can also start things from scratch, such as writing a book about apples or write me a script for a podcast talking about tornadoes. In terms of scripting and brainstorming, this platform is a great place to come to *** not only free, but it's pretty good at what it does. Now, you have your script, you have your audio. What about footage? One place that you could get images from is Mid journey. This is another AI platform. Nowadays, they also do short snippets of videos. I will talk about videos in a bit, but if you wanted to do voiceover over images, then this is the place to come. This is not a free tool. I believe even if you do want to use the free version, there's a lot of limitations. But as you can see, these are some samples that you can make with this y tool. Once you're done making them here, you can export them into ElevenLabs, do your voiceovers and make some content with them. Now, what if you want to make videos? Runway AI is a great platform for videos. There are others out there, but this one is probably the most easiest to use. The way this works is that just like any other AI platform, you give it a prompt, as you can see here. You can play around with angles, with the mood, the characters, keep characters stable, and just do a bunch of different things. This, once again, if you're using the free version, there are some limits. You could upgrade to a better version, but if you don't plan on doing this that often, you don't really need to buy a full account. ChaGPT also does generate images for you. If you didn't want to pay for Mid Journey or DaVinci, you can just come back to Chat GIP Team. This way, you're combining scripting, imaging, and video capabilities with your already made audios. This way, you have the option to generate different of content and not fully rely on audio only content. Now that we have all of these tools at our disposal, let's talk about how you can monetize these content if you choose to. I will see you guys in the next lesson. 44. Monetization and Next Steps: There are a lot of ways that you can monetize with ElevenLabs just as you can with other AI tools. So apart from being able to make your YouTube channel and monetize through that platform by making content with ElevenLabs, you can also join in on ElevenLabs own Creator program. So when we go to the Voices Library, there are some that charge you per minute. And you can see there are not that many, but they do exist. Essentially, what you can do when you make your own voice is that you would be getting paid passively after you have uploaded a high quality audio on 11 labs. For example, this audio is I go over it, $0.02 for me to use. Dude, if riding waves is my religion, then tacos are straight up communion, bro. So that's an example. You can see it's very high quality, very clear, and it has the perfect voice for this category. So that's one way that you can monetize ElevenLabs creations. On that note, they do have a carriers page as well, where you get to be one of those specialists that you can use your services from in production. Recall when we went over here, we were getting, if we were to use it, charge $2 per minute and all these pricing. These would then be made by real humans as we discussed. You could get qualified to be one of those people. Simply by going into the slash careers page, then open positions, you can see how much stuff they have available. Now, those are full time jobs, but if you go in the transcription slash subtitling, you can see that this is a freelance thing. So whenever there's a demand for your service, you're going to get reached out to and then record an audio with your voice for someone else. You can see it's even remote, and this is all within this platform. You should definitely check it out if you plan on making money with ElevenLabs. Other way that you could do it is make content online, such as YouTube videos, Instagram videos, Tik Tok, or whatever, using ElevenLabs as audio. This will save the time for you to record everything with your own voice and you get to use these really quick and accurate audio generations to build your content. For example, you could cover scary natural disasters like we did with one of our projects. You can do a biography overview where you tell the story of famous people. There are so many things out there that you can do. But the role of ElevenLabs in that is that it's just the audio. You still have to work really hard to advertise your content, get it out there, have people watching it, be engaged, and all of that stuff. Another thing that you could do is freelance with ElevenLabs. Basically, you could go on platforms like Fiber, Upwork, and there's so many out there nowadays and offer the service of making AI audios for a client. If someone doesn't want to use their own audio, they can come to you so that you could accurately make them and audio. The same way we altered the voices for a scary podcast, a advertisement, a regular podcast, you would be doing the same thing, but based on the client's need. Those are just some really quick ways that you can monetize with 11 labs. Just like any other AI tools, there's going to be a lot of demand for this type of content and work. Keep an eye out and make sure that you're in the right communities so that you're aware of all of the updates and all of the cool tricks that you can do with this platform. Redit has a lot of them, and then you can go to maybe Facebook, Discord and just keep on practicing making audios and learn new ways to make them better. I hope you guys enjoy this course and that you continue using this really cool program for your own content. Hope to see you guys soon. 45. Class Project: Create Your Own Audio/Visual Content: Now it's your turn to take what you have learned and create your own project using ElevenLabs. For your class project, I would like you guys to create either an audio or a visual content using all the tools that we have learned. You can also combine it to where you have the visuals made and then combined with the audio that you made. You plenty of freedom here. You can create a short video, an audiobook, a podcast, lip sync video, and anything else that you want to do. Start by deciding what you want to make and who it's for. Then think about which of the many tools can you use to create this project. You can choose or design an appropriate voice for your project and see how that could fit with your idea. Could just be an audio project. But if you want to take it a step further, you can add some visual content made from the creative tools. We learned how to generate images, videos, and how we can make a full production using flows. So choose the tools that genuinely help with your idea. When you're finished, you can upload the project to the class project library. Depending on what you have made, you can include the full content, screenshots of it, and alongside that, you can provide an explanation talking about your process, your idea, some challenges you face, and what you were curious about. And remember, when using AI to generate voices, you want to do this responsibly. Especially if you're cloning other voices or you're creating content based on real people. Looking forward to seeing what you guys create. 46. Congratulations! What’s Next?: Mm hm. Congratulations on finishing the ElevenLabs AI master class. We have covered a lot in this course, but by now, you know that 11 labs is so much more than just text-to-speech. The important thing now is to keep experimenting with the tools. You don't have to use every single tool in your project, but you can see which tools would help improve your workflow or help you create better content. Or even larger workflows, you can combine 11 labs with other AI tools out there. This could be for whether you're working for a business with a single client or for your own personal project. If you haven't already, be sure to upload your project to the class Project library. I would love to see how you guys use the tools and how you combine them to generate your ideas. And if you enjoy the course, feel free to leave us a review as it really helps us improve our content. Thank you for joining me. I hope this course have left you with a lot of ideas on how you could use ElevenLabs for your own projects. Keep creating, and I'll see you guys in our next class.