CapCut & Google Veo 3: Cinematic AI Video Mastery

1. Introduction: What if you can create a short film containing visuals, voiceover, editing, everything without picking up a camera. Sounds crazy, right? But with AI, it's not just possible. It's powerful and blueprint and welcome to AI filmmaking with Google Flow, VOT gemini and Capco. Is a beginner friendly class that shows you how to turn simple prompts into cinematic stories. In this course, we are not just learning tools. We are making a real film. I'll guide you step by step as you use Google's newest AI tools to generate scenes. We find your story with Gemini, voice it over with AI, and edit it all together with cap code. This course is perfect for YouTubers who want to develop their visual storytelling. Content creators looking to make cinematic AI power videos, filmmakers experimenting with new creative AI workflows, educators, marketers, and generally anyone with a story to tell. You don't need a camera. You don't need a them to you just need an idea. And the best part, you walk away with your very own short film, completely AI generated, beautifully edited, and uniquely used. So if you're ready to learn the future of storytelling, begin this class right now, and let's make something cinematic together. 2. Project: Creating a Short Film Using Google Flow: This is the projects you are going to be submitting after taking this course. And the goal of these projects is to bring a story to life using AI Filmmaking tools that you're going to learn in this course. So these are the three topics. You have to choose one of these three topics, a lost boy in the desert, the story of Noah and the arc or memories in a mirror. And you can see the topic description right here. Use Gemini to generate the script, three narrative voice over script, emphasis on narrative voiceover script, right? Then it generates images for the characters and environments using the tools that you're going to be learning in this course. All right. Then you go ahead to create the visuals in View three. That's the video scenes. You create them in View three. You create your voiceover. Then after creating your voiceover, you go ahead to combine the visuals, audio, add transitions, add ambient sound and background music when necessary using cap codes. Once you're done, you upload your video to YouTube. You can make it an unlisted video or you can make it a public video, then you past the link. You go to the project section and you submit the link. Also, you submit a screenshot of your Gemini generated script. There should be a part where you can upload an image. If you don't find it, you can as well just add a screenshot of your script at the end of the video. You are creating your video, make sure you focus on clarity and emotion over complexity. Your story needs to be simple. You don't need to use complex sames or complex scripts. Make sure it's not too long, okay? So your video should be between 30 seconds to 2 minutes. 1 minute is a sweet spot. 30 seconds be two shots based on your script, it's up to you, but just don't make it past the 2 minutes mark. 1 minute is even enough, okay? And make sure you choose music that fits the tone of your narration. Those be afraid to let silence piercing and ambient sounds do parts of the storytelling. It's not all about talking, talking, narrating, narrating. There are some parts where silence will actually help the video. And lastly, make sure you experiment. You can always tweak and refine. You might not get the best video from Beauty at the first trial, but if you keep trying tweaking your proms, just regenerating, you're actually going to get something that you can work with. Alright, I wish you all the best in this course. Please make sure to leave a good review if you learn something from this class and make sure you submit your class projects. These two things are very important. See 3. Accessing Google Flow: So if you want to access Googleflow, what you need to do is to visit the URL, labs doggle.com. Okay? So I'm just going to type labs dot Google. Actually, it's just labs dot Google. And it will bring me up to this page. So I'm right here on this page, and the fourth featured AI to here is flow. So what I can just do here is to click Create with flow in order to access it. Okay, so it's as simple as this. You can also just search Google flow in your browser on Google, and it will show you the link where you know the link you can click to access it, right? So it has loaded up, and this is Google flow. See here that it requires a Google AI subscription. So as at this moment, Google Flow is not free yet. It requires a Google AI subscription, as you can see below. So I'm just going to click Create With Flow and sign it with my Google account. So I just signed in to Google flow using my Google account, and you can see, I can't do anything, I can't generate anything if I don't get a subscription. So what I can do now is just click this button here in order to get this subscription, right? So I will just close all of this now. And yeah, so you can see the two subscription tiers that you need to be on that you can either be on, you know, that access Google flow include the Google AI Pro and the Google AI Ultra plan. Okay? So this first one year is the Google AI Pro plan, and it costs $20 per month. So when you subscribe to the Google AI Pro plan, you're going to get 1,000 monthly AI credits that you can use across Google Flow and Risk. Alright, so these credits are just among the other benefits that are included in this subscription. Okay? You can see others like the Gemini emiPro the Notebook EnPro too, and others Gem Line JMU Docs and 2 terabytes of storage. So if you want to see the other plan, which is the Google AI O Trap plan, all you need to do is to click the CO plans and to bring up that page that contains the Google AIO TRS subscription. Place side by side with the Google AI Pro subscription, as you can see right now. So the Google AIO Trap plan is actually pricey. It costs $250 per month. So this one is basically for the movie industry. Let me put it like that. So for companies or YouTube, Brazil, that are really big and they need access to this Google AI to for commercial purposes. That's why it's price like this. So you can see that when you do this subscription, you're going to be getting 12,500 monthly AI credits that you can use across flow and risk. Don't worry if you don't understand what credits are. I'm going to be showing you later what credits are and how they are consumed whenever you generate videos using F two or three. Okay? So this is the comparison. This is just what you need to know. When you get the Google AI PRO plan, you're going to be get 1,000 monthly AI credits. This is what we are concerned about here. But if you get the AI ultra plan, you are going to be getting 12,500 monthly AI credits. Also, if you subscribe to the AI Pro plan, you won't be able to use gregens to video. So gogent video is a feature in Google Flu that allows you to add multiple images, okay, multiple ingredients. To your video. So let's say you want to make a movie that contains, let me say, five different characters. And I want to make these characters consistent across everything. Ingugens video allows you to upload all of these media files and use them to create your videos. But the AIP actually has a feature, which is the frames video. I'm actually going to be talking about this later, but I just want to explain it a little bit now so that you understand what you're going to be getting and you'll be able to make the right subscription choice. Okay? So the frames video allows you to upload an image that is going to be specifically used to start the video or to end the video. And you can also use that image, that particular image later, you know, throughout the video. It works. So it's almost like the gagent video. It's just that you will only be allowed to upload two images, okay, two images. Do you understand. Lastly, there are some camera movements that you might not be getting in the Google AI Pro subscription. When you are making videos with Google Flu, it actually has a feature where you can choose a specific camera movement, whether it's panning to the left or panning to the right or coming from up to down, you know, those camera movements, you can select any of them for a particular scene in your video. So you're going to be getting access to both of them in the AIP subscription and the AI ultra subscription, that I believe the AI ultra has more camera movements than the AIP subscription. Altogether, the AI Pro subscription is best for small creators, okay? It's perfect for those who just want to try out the features of Google Flu without breaking the bank. Wow, the AI Ultra plan is meant for big creators, big YouTubers, the movie industry generally. So if you can afford it, I would advise you go for the Google AI ultra subscription. It's totally worth it. But if you feel like $250 per month is a lot, then you can just go for the Google AI Pro. Also, if you don't have the money right now to actually subscribe to the Google AI P plan, you can see here, you can get it for free for a month. Okay? So in order to take this course, you need to have access to either the AIP plan or the Google AI UTA plan. But if you don't have the money, all you need to do is just to click this button, which allows you to access Google AI Pro for free. All you need to do is to add your card and you won't be charged, but you'll just be given a free tier o it's going to be for just a month, but it's enough for you to practice everything that I will teach you in this course. So that means if you're starting this course right now and you are getting your subscription, you need to complete this course within a month. It's not even a load, actually. You can do it, and I'm advising that you do it, okay? So what I'm going to do now is to add my card. So you can see now, I just added my card to it, and so what I need to do now is to click this subscribe button, and I'm going to get access to Google AI Pro plan for free for a month, right? So so you can see all the featured benefits that I've gotten access to just by adding my card to my Google account as simple as that. Now, welcome to Google Flow. In the next lessons, I'm going to be showing you exactly what you need to do or the steps you need to take in order to create cinematic videos. But what I need you to do right now is to go to labstogog.com, selects Googleflo and get a subscription. Either you get the Google AI P one month free try or you just do the normal subscription or you just go for the AI ultra plan. Okay, whichever one you choose, it's up to you. I already told you the benefit and you know what to do. 4. Get Veo 3 for FREE for 4 months!: Before you continue this course, I need to share this updates with you. So my free trial on this particular account is nearing its end already, and guess what I just saw. Look at this button right here. It just appeared out of nowhere. And if I come to another account that I just did this free trial on, you'll see that the button to invite a friend is not there. So when you do your free trial, look out for this button. You might not see it at the beginning of your trial. You might not see it at all, but towards the end of your trial. You are going to see this button. Now, what happens if you click this button? Let me show you. You can see what is here. Friends will get a four month trial or Google AIP when they subscribe with your Invite link. Now what this means is that if you copy this link and you paste it in another tab or in another browser or in another window, and then you activate that free trial using this particular link, you are going to get four months of Google AIP for free. Okay? So instead of just getting a one month free trial, if you use your own referral link to get another free trial, you will get four months of the Google AI P plan. And that's what I did right here. You can see that this is a new account. This is my old one. I just copied the link, my invite link, and then I got another card and used it to get this four months free trial. So every single month for four months, you're going to be getting 1,000 AI credits for use in Google Flow. And you're also going to have access to every other Google One feature and every other Google AI Pro plan feature. So take note of this. This is how you can get Google AI Pro plan for free for four months. And if this invite a friend button is still here after the four months, then you can redeem another free trial using another card, and then you keep on doing it over and over and over again. That's where you never need to pay for the AIP plan, so enjoy the court. 5. Flow Interface Walkthrough: Welcome to Google Flow. This is the interface that you're going to see when you initially open Google flow for the first time. Okay? So there are different places that you can actually click around here and I'm going to show you around this Google flow so that you can navigate around it with ease. All right. So this flow here, like I say, if you click it, it's simply bring you the home page, this homepage. So let's assume you visited Uh, you click this button or you click some other button here, and you just want to go back. Let's go straight back to the homepage. You can easily just click this flowy and it's going to bring you back to the flow homepage. So if you want to create a new project, you can see this big button year. This is what you are going to click to start a new project in Google Flow. So this button year is for flow TV. Flo TV is like the YouTube of Google Flow. So if you click this button here, you're going to be able to watch videos that other creators have created with Google Flow. Okay? So if you need inspiration or you want to see what you can actually achieve using Google Flow, you can easily just click this button and visit it, visit, you know, the flow TV. I'm also going to be talking about this in another lesson, alright? This bots in year, this icon year is simply to adjust the volume. Okay? So you can slide this down to reduce the volume to the barest minimum or slide it up to increase it to the highest. Okay? So this icon year is the FAQ icon, frequently asked questions. If you have any question that you don't have a certain answer to, you can easily just check this FAQ to see if it has been answered before. So commonly asked questions are always, in this FAQ section. Now, we have this button here, these three dots here. If you click it, you are going to see this watch flout TV again, which you can also just access from here straight up, and you see this button here, this icon, you have to learn flow. So if you click this button, it's going to redirect you to a very short YouTube video that just explains how flow works. And you can see this one's self explanatory. I don't think you ever need to use these two buttons. But, this one is for sending up feedback. If you have any problem it's Google flow or you want to report any legal issue. You can click it out you can click this button for legal issues, and this one to send up feedback. All right. So if you click this P icon, you are going to see the number of credits that you have remaining. You can say that I just created this account, so I have 1,000 AI credits. It's an AI Pro subscription. If it's an AI ultra subscription, I'll be having 12,500 credits right here. Okay? And you can just click this button here to upgrade. If you want to upgrade to the AI ultra version, you can just click this button, right? This is my library button. I'm going to be showing you in a bit in my other accounts because I don't have any Fjet here, you're not going to be seeing anything. So let's skip that and go to this managed membership. If you want to manage your subscription, if you want to cancel your subscription before the due date. So this one is very important. If you are sure that you don't want to do a subscription, make sure that you before the end of the 30 days trial, the one month's trial, make sure you come to this managed membership and you cancel your subscription. You click the managed membership button, this is what you're going to be met with, okay? So if your credit is finished and you have a subscription, you want to upgrade without using the AIO Tra plan. You can easily just upgrade to the $25 per month subscription, the $50 per month subscription. You can see if you click this, you're going to see how many credits that is going to come with. Okay? So if you want to cancel your membership, look at it at the bottom here, just click this and it's going to cancel your members. Make sure that you do this before the end of your tryout period. If you don't do it, Google is going to charge you for a year's subscription, okay? You're going to charge you for I think it to charge you for a year out right, not for a month, for a years worth of either to charge you for a years worth of subscription or for a month's subscription. I don't know which one, but if you're not ready to subscribe, then make sure you cancel your membership. If you know you're going to subscribe, then, it doesn't matter. You don't need to cancel it. That's it if you want to sign out of your account, you can easily click this button right here. Let me go to my account so I can show you what your dashboard will look like when you actually have projects on it. You can see now, let me just click this flow button. Let's say that I have several projects here. And if I scroll like this, if I just scroll, I can easily just go through the projects that I've created. And each project has a name. The date that I created it, I see May 29, but I come to this one June, so that's second of June, okay? And if you want to see everything, if you're not comfortable with this scrolling cylinder, okay, this scrool Snida form as, if you want to see all your projects easily, you can easily just click this grid button, this grid icon, yeah. We click this grid icon. It's going to display the projects, you know, this side by side. So you can easily just select one you want to select. And if you want to rename your projects, you can easily just click this Spencer icon here, and then you rename it. Okay? So let me just put first projects here and just lick this bolds in see, that's how to rename your project in Google flow. Alright, so let's visit the library now. If I click this and I click my library here, you're going to see all the different generations that you have created over time. So there's a difference between this library section and the project site. Okay, the project section, which is the homepage, there's a difference. Now, in the project section or the homepage, you're only going to be seeing the singular projects that you have created. In this library section, you are going to be seeing each and everything that you have generated in Google Flow. Okay, these are individual scenes that are created, and every scene in Google Flow is 8 seconds in length. You can see 8 seconds, 8 seconds, everything. These are scenes that I've created, and they are in this library section. So if you want to access a particular scene, a particular video that you generated, not a whole project that you need to come to this library section. Let's say you created a particular video sometime ago, you know, just the text video, you created it probably two days ago, you want to reuse you want to download it and use it as an asset in a particular video that you're editing in probably cap codes or, you know, some other software, video editing software. And instead of opening a full project and trying to download a particular say, you need to come to my library section. Come to your library, and then you simply find, you scroll, and you can also type here, too, okay? So if you type, let me just type Spider Man. Okay, then enter. So where you search is going to search through all the prompts that you fed Google Flu with. If I search for Spider Man now, it's going to bring out all the video generations, all the things where I used the word Spider Man in the prompt. Okay, so if you don't remove, you have lots of projects, a lot video generations or a lot of scenes that you've created, and you don't want to start schooling to find a particular one. You can easily just search, you know, if you remember the prompt you use, you can just search, you know, search for a particular word that you used in the prompt, and you see you'll be able to find, you know, it just filter out the rest of the video generations that's not include that particular word in the prompts. You can see how easy it is, right? So see now wine, the video effects. Okay? So if you click this drop down, you're going to see a drop down list of all Google AI products, the music effects, image effects, which is, their self expanatory just. But in this course, we're going to be focusing mainly on this video effecs which is View three and Video two, Google F. 6. Exploring Flow TV: Let's explore flout TV and see what's actually behind this button. So just click this button here and let's see what we're going to see. So right here, you can see the flu TV, see what it looks like. You can see this video was generated using View three or two. I'm not sure which one but they have it changes. So after watching a particular video, it's going to auto switch to another video. Okay? You can see. So here you can select all channels and you can view all the channels that are available in the float series. So you'll see different channels Spectromatic, Micro Vas, ultra white, snouts and about. You can select a particular channel that you want to watch. Let me click this Mico Vas I see the videos that they have. So let's wait for something to play. Kind of having issues with my Network, so I won't waste time here. If you want to change it from the Lights box view, that's that cylindrical view. You can easily just click this great view, and it's going to show you it's going to show you all the videos like this. So you can see the videos that are on this channel by just hovering your curso over the videos. You can see. And if you want to pause, you want to play, it's obvious if you want to loop a channel, I mean, if you want to be watching all the videos in the Channel without because if you don't loop a channel, once it's done watching you're playing all the videos in the Pasqua Chanel, it's going to move into another. If you want to keep watching only the videos in a particular channel, you can easily just click this button and it's going to loop the channel. Oh, this button here is going to I don't know the difference between these two buttons, but they are quite similar, because if you click this button, it's going to take you back to the all channel section. Okay? So before I click that, you can see here, I say this up and down button. You can click here to change a channel, to change another channel. So you see I've switched to the ultra white channel. But click again, it's going to switch to another channel, okay? You can see go to previous channel, go to previous channel. And that's how you navigate through the flow TV. So let me click this view right now. I see that has brought me back to this place here. So if you don't have any particular channel, any particular favorite channel that you watch, you can easily just click this Shuffle and it's going to bring you up to this stage where it's going to just be playing videos from different channels, right. So let me come back to this section now and show you here this pace here. You visit this short film section here, you're going to see different short films that have been created by fellow AI movie creators like. Okay? So there are different wonderful short movies here. If you click this button, you're going to go to another out the next one and onto the next this electric pink is actually a very wonderful movie. If you watch this movie, I'm very, very sure you are going to be old. Okay? It's going to show you how much you can actually do with Google flow. It's a wonderful movie, right? You can click here to visit the Creators Instagram page. Click here to visit the Creators X page, you can easily just click here to watch the movie. All right, so it's as simple as that. This is how you navigate the flow TV section. You can see this button here to create its flow. I want to go back to the flow homepage where you can easily just create your own video with flow. Just click this bots in E so take you back to the homepage, okay? So this bots in year is for joining the Flow Discord channel. Just click it if you have a Discord account and you can join it if you want. This one is also for About one's self explanatory. So that's how you navigate the flow CV section. And also, don't forget you can actually just search for any flow movie that you've actually. You know that, Okay, so you can search for any thing of flu TV. So if I tap pink now, let me see if I will get results for pink. You see? There are different results for pink. Search in this space. I'm not sure. Google Flu actually has a vast collection of videos here yet. I'm not sure, but you're surely going to see different videos that you can watch right here, right. 7. Veo 2 vs. Veo 3: Key Differences: Now let's proceed by clicking this new projects button right here. Okay? So once you click that button, this is what you're going to be met with. This is what the Google flow interface looks like when you want to start creating. Okay? So you have this drop down here. If you tap it, you're going to see text video, which is simply just typing in a prompt and letting Google generate a video from it, or let's two or generate a video from that text. Okay? Now, D framed video allows you to add a starting scene, okay? So if you have an image and you want the particular scene of your video to start with that image, then you add the image. You simply click the plus icon right here and you can upload or generate an image, okay? So here, you can generate an image using imaging or you can upload an image from your gallery. Or file manager, right? So here, you click this one. If you click this one right here and you upload an image, the image is going to be added to the ending parts of the scene you understanding, right? So the last option, which is the Inglgens video, I mentioned this earlier. This one allows you to add multiple media files that you're going to be using or that you want to use in your scene, in your video scene. And it's only available to Google AI ultra subscribers only. So if you want to have access to the ingredients video feature, you need to upgrade to the AI ultra clan. So right here, we have this icon, and if you click it, we're going to see different camera movements. You can see the Dolly in Dolly Out, Jeep down, Jep up, and you can select any of these camera movement to be used in the particular scene that you are creating. So if you don't know the meaning of any of these camera movements, you don't need to stretch yourself. Hover your co over the particular camera movement, and you will see what it looks like. I said, This is what the dolly in looks like. And if you ho this, you're going to see the dolly out. This is what it looks like. If you look at this, if you hover over this tills down, you see this is what it looks like. So if you select a particular camera movement, we select it here. And you type in a text here, it's going to Jen darken using this particular camera movement, okay? And if you want to cancel it, you can simply just click this Cancel. You click the icon, and it's going to Cancel, okay? And now, for this part here, this certain spots. If you click it, you are going to see these two things, the output spa prompt and the model. This housefut spa prompt simply means the number of videos or the number of scenes that are going to be generated each time you generate a video. So by default, where you type in a text, and you generate the video by clicking this button, it's going to generate two different scenes. Okay, two different scenes. So you can choose the one that you like from the two sins and note that whenever you generate just one sen, you see, each generation uses ten credits based on your current setens. So if you leave it in these two output spa prompts, whenever you generate a video from text prompts, you're going to be spending 20 credits because you're going to be generating two different videos. So I recommend that you change it to one Okay, change the outputs per prompt to one. So that just one video is going to be generated whenever you hit this generates button. Okay? What this means is that you're going to be spending just ten credits whenever you generate a video. If you're limiting two, you're going to be spending 20 credits per generation. Because we're going to generate two separate sins. Do you understand? If you change it to four, they're going to be spending 40 credits a generation. Now, the number of credits that you are going to be spending actually depends on the mode two. Now, if you click this drop down here, you're going to see View two and View three, okay? So the View two has two different settings, okay? So it has the first and it t the quality. I view three just has the highest quality. So this View two fast uses just ten credits per scene generated, okay? And if you select this quality now, it's uses 100 credits, 100 credits per scene generated. You understand? That means if you leave the house foods per prompt, this is going to use up 400 credits because it's going to be generating four different videos at the same time. Now, this View three, which is the highest quality, also uses 100 credits. So I advise you change it to one outpos per prompt, okay? Just one aspos per prompt is okay, so that whenever you generate a video, if you generate a video using the first view two, it's going to just use ten credits. I've used View three, it's going to use just 100 credits. Now, the main difference is just the quality. Obviously, View three has the highest quality and it also comes with audio. So whenever you generate a video with View three, you can make your characters talk. You can make the AI add sound effects to your video. Okay? It's only available in View three. You can see it's experimental audio. If you choose View two, you're going to get high quality video, okay, better outputs, but without sound. The same thing with View two. If you choose the View too fast, rather, if you choose this view too fast, you're going to get a good output, okay? You're going to get a good output, but it's not going to co beat all do. Now, don't consider the fact that you're spending just ten credits on VotooFast. You can actually generate decent videos using View Too fast. So I encourage you to try it. We're going to be using it also. View two actually works perfectly if you want to generate views, just random sins to fill up a part in your video. Like maybe you just want to add the vial of I said, that's started like a joe footage over the hills, over mountains, you know, just to fill in a particular part of your video. You can use View two for that. It works very fine. It works very, very finely. Most of the videos that are generated using Google Flu as being this View two fast, okay? So it's actually very economical. You just spend ten credits per generation. But if you actually just want to take it notch up, you can use the quality version of View two, okay? You can use if you have enough credits, you just want to try it out. And if you want to generate with audio, Okay, you can use this view three, which is the highest quality, and thankfully, it just uses the same hundred credits as the quality View two. Alright? So that's the difference between View two, first quality and View three. So I encourage you to just play around with this sentence now. Not just play around. Just do what I've told you to do. Change your outpos by prompt one, or you can even use two. To is manageable if you are using View two, the first version of view two, okay, you can generate two videos at the same time, and I'm going to believe in it. So if I want to use view two, and quality of three. I'm going to be changing it to just one house bolt per prompt, right? So we have this but in here, this icon here, so clay proms. If you click it, it's going to clear the proms. Let's say you type something and you want to change it to something else and you don't just want to use Control A and backspace, that bolt in year will help you to clear everything at just one click. Alright. Right here, so you can see this icon here to change it from the column view to the grid view. Okay? So when we generate the video, you're going to see what it looks like exactly. So make sure you complete this course. I promise you're going to enjoy the whole process, right? So that's it, and I'll see you in the next lesson. 8. Text-to-Video Generation: Now, I'm sure this is the part that you all have been waiting for and I'm going to get right into it. Okay? So we're going to generate a simple video using View two. We're going to be using this takes video mode, right? And we're going to be making sure this isn't the view too fast. I'm going to leave it at two outputs per prompt. Now, I'm just going to type in a simple prompt. Okay? I'm going to type in a simple prompts like a boy walking down a lonely desert trail. Simple. A very, very simple prompt. Now let me change it to this column view. I prefer this column view. But if you prefer this grid view, you can easily just switch to the Grid view. And here, that's it. While this is generating, let me just explain what this part means. You can see the prompt that I used there just beneath the generations and you can see the mode I used, which is the first mode, the view too fast. And right here, if you click this, you can reuse the prompt if you want. Say you're not satisfied with the generations right here, and you just want to tweak the prompt. You can simply just tap this and it's going to prefill this textbooks with the initial prompts that I use. Now you can decide to edit this to your taste, add what you want, or subtract what you want to subtract. So here, you're going to see the option to delete your outpoT delete the generation or to flag the outspos if the outpose doesn't sit well with you, right? So now, the first one has been generated. Let's split it and see what it looks like. Well, do I don't like this, well, because why does a desert have grass? Why does a desert have plants all around? But this is wrong, I guess. A desert is supposed to be dry. It's supposed to be just no plants, nothing, dry lands and hills and, you know, sand dunes and all that. It's not supposed to have grasses around plants around. So, well, this is the generation, first of all, will working on this later I don't know. I think the network is bad, network connection is bad, so that's why this one has not generated. Okay? So if you look at this, you can see this option. You have different options here. You have this option to make this full screen. You have the option to download. And if you click these three buttons here, you see the same thing to flag the output or to delete this particular output. Right? And if you're creating a long video using Google Flow, you have this option to add it to a SN. Okay? That's why Google flow exists so that you can create a video that consists of different things. We're going to be talking about this in a different lesson. So for now, just know that this exists. And if you want to download your video, you can easily just click this button. Once you click this icon, you see the option to download it as an animated GIF, so seven CP original size 720 CP or an upskill 18 CP video. So this is the highest quality. If you click this, it's going to first of all, upscale the video. When it's done upscaling, you're going to see an option to download it. All right. You'll see an option where you can just click it. The video will start downloading. Now, so this is a very basic prompt. There's nothing here, basically. We just said a boy walking down a lonely desert street. I didn't actually add much information for V two to use, you know, to create a quality video for us. So that's why your prompting is very, very important. The prompt you feed View two is very, very important. You need to feed it with as much information as you can feed it without overdoing, okay? So you need to describe what you want to see. Imagine in your mind what kind of video? What do you want to see in this scene? What mood do you want this scene to give to the viewer what vibe you wanted to give in the weather conditions. Just imagine it. And then once you are done imagining, you come to this text video here, and then you type, you describe what you imagined in your head, okay? Exactly what you imagined. That's what you're going to describe to it so that you can get better prompt. Now, there's a particular prompt format that I found online, and it's actually very, very helpful. So I'm going to just pay sits right here, I'm going to explain it to you. You can see that it has finished upscaling. And if you want to download the video, you can easily just click this download button. Is going to download this right to your device, or you can dismiss it. Now, you can see these prompts here. You can see the way it is arranged and how it's described. First of all, the prompts just describes basically what the video should be about. Okay? You can see it's a charcoal sketch animation with a raw and expressive aesthetic. And yeah, we have the time of the day. Have the color and mood. We have the camera lens and position. We have the character description. So if your video has a character in it, that's when you're going to be finishing this character description. But if your video doesn't have a character in it, let's say you just want to generate a Bal of a drone footage, you know, off range of mountains, a mountain range or something like that, then you don't need this character part here. Okay, you might not need it, so you can just use it to describe the mountains. And then this statis short and environmental details, environmental details. This is where you describe the background, the surroundings, where your character is going to be. I'm just going to edit this prompts to match this particular video that I created earlier. Now look at the prompts that I just cooked. You can see how different it is from the initial prompt that I used to create the sin here. You can see how much more detailed it is. You see that I follow the templates that I pasted here. I'm going to add these templates to the resources file that you're going to be getting after completing this course, right? So you can also check it. Check the project section, and you're going to be able to download it. I'm going to include different prompts that you can just tweaks you know, tweak some things and let's get what you want from it. You understand so this is the prompts now. You can see that I added the color and mood, added the camera lens and position, the character's description, you can see how descriptive I made it. Also the environmental details, okay, time of the day and everything, look at the prompts, right? So what I'm going to do right now is to feasible, make sure that it is in view too fast, and then I'm just going to use two output spa prompts, right? So I'm going to send this same fasib copy this right just in case. I'm going to sell this, okay? So I need to refresh just you see that arrow that popped up on my screen, that simply means you need to refresh your page. And this only happens if you've left the tab unattended to for quite a long time. Do you understand me? Now, you can see that as I just refresh this page. It's already showing me that there's a flow updates already. So let's read it together. The excitement for our latest model, View three has been overwhelming. Okay, let me just zoom this in, okay? We're excited to bring in a new VOT first option for text to video that costs only 20 credits. Wow. Wow. So I just read through everything and all I can say is Google is working. So I'm going to explain what's happening here in the bits. So let me just start to get started and see I'll pass this here. So if I click this icon here, the Setans icon, and I click the Model, you can see now that there's a new model right here, DVT First. And it costs only 20 credits. So you can say that just while I was creating this course, I told you earlier that V three costs 100 credits. Okay? And VT is the only one, the only model that produces videos with sounds. But now, Google has made an update and added Vw three fast. You know, we have VT fast normally. So they added Vw three fast and also comes with audio for only 20 credits I'm going to use this. Oh, before I use that, I want to use this view too first. Okay, so we're going to use them one by one. So let me just select this view too fast and show you what it's going to look like using this prompt, right? So this is going to use just thin creative as you can see right here. So I'm going to hit this button here, this icon writing and wait for it to build my scene for me. As you can see, our view, two videos have been generated, and they were generated fast Pytended. So, let's play this right now and see what it looks like. Let's put this first one first. Let's look at the prompt again. Look at the prompt. Look at the prompt. Okay? Well, I did not do this, this part because I told it to rise up, so slowly rise up till it becomes an aerial shot straining from the top. Let's play the second one now. Okay. So you can see the two videos that we generated and look at the former one. Look at the former one we generated with a simple prompt. You can see that there's so much difference. You can see the differences between this one and this one. You see that I added I told it to add a cap to add the red shirt and added everything, the water bottle too. And this is where I used a boy walking down a lonely dessert trail. You can see that it added plants or necessary plants to this video. In this video, I specifically told it to remove the plants, even though I have some baby cctide. But it's alright. So I'm going to piece this symptoms right now. I'm going to piece the symptoms right now. Here. And here, I'm going to add aerial drone a high AEO drone shot. Great. Now I'm going to change the model to View two quality. Okay? So I'm going to use the quality mode. I'm going to change the outputs platform to one, so I'll just use only 100 credits, and then I'm going to send this and see what it gives me. Alright, so the View two quality is finally done. Let's play this and see what it looks like. What? Okay, so okay, okay, okay. Let me let me play this again. Let me play this again. Nah, no, no, no, no, no, I don't like this. I do not like this. I do not like this. You can see what's put at the beginning. This AI is still in the development stage. I'll just put it like that, okay? But it still creates great stuff. So the last one I'm going to be using now is the View two, the view three, right. I'm going to be using the first one first. Okay? So let me use the first one. So now I am using the first View three, but just one outspot per prompt. So let's see if this one is going to be better than the view, the View two quality, okay? Alright, so the VT first is done. Let me play this now. Ooh Ooh. Okay, okay, okay. So you can see the difference. Now, you can see the difference right now. View three is so, so, so much better than View two. That area shows that I needed. Okay? The one that would be like a drone, I see if the video was being taken from a drone. I did not get it here at all. So we went on to use the View two quality, and we got rubbish. We got rubbish here. And the last one now, which is the view three first, which costs us just 20 credits. You can see what is created. You say, the landscape is even so much is much more hyperrealistic. There are no plants around. I don't see any cacti here, but I could see some cacti in this particular video right here from view two. And it's got the aerial footage. You can see? You see. Beautiful. Beautiful. So I have just shown you how important it is to use a very detailed prompt. You can see the kind of outputs we got here with View three and the detailed prompt and the kind of outputs we got here with View two and the lazy prompt. Let's see the difference. So whenever you are creating videos using flow, make sure that you put some efforts, put some efforts into crafting your prompt. Imagine what you want your video to look like in your head and try as much as possible to type out a description of what you can see in your head. This formula I have just given you works for me every single time. So I just imagine the kind of video that I want to create in my head, and then I start typing it out, trying to, you know, the tiniest details down is the most minute details. I try to express myself. I try to convert my imagination into words. So if you follow this formula, your videos will always be great. Well, they will be great if you use VT. But as I said earlier, View two should not be underestimated. I'm still going to be showing you later in this course. We're going to be working on a video, and I'm going to be building some parts of it using View two. View two is still amazing, but for very, very detailed prompts, for very detailed videos, you might just want to stick with View three. So this is it and make sure to check the project section for the prompt spark, okay? So you're going to get this exact prompt that I used in this video so that you can edit it and use it to create your own video. 9. Text-to-Image Creation: Generating Characters and Environments: In this lesson, I'm going to be walking you through the process of text to image generation. So I'm going to be showing you how you can generate characters and environments for your video. There are different ways you can generate images. I'm sure it's not a new t to you now, but there are ways that you can actually create stunning realistic images for your videos. Okay? So the first one we're going to be using is Google Gemini. I say, it's just geminigle.com. Okay? And if you already added your card to your Google account and you got the free one month subscription, you would have access to this model. Okay? This pro model, 2.5 pro. Okay? So here we can easily just ask it, create an image of an Ara tuni supero superhero reads an eye patch on one eye and a green suit. So this is a very simple prompt. Okay? So I'm trying to create my own superhero, okay, an image of my own superhero that I can use in View three. So let me just send this now and see what happens. So while this is creating, what I'm going to do now is copy this same prompt. Put copy the same prompt. I can see that it I started generating the image. So it's done already. Tous first. And you can see the tip right here. You can ask Gemini to refine the images. It's image generation in Gemini, your imagination is the limit. If what you see doesn't quite match with what you had in mind, try adding more details to the prompt. The more specific you are, the better Gemini can create images that reflect your vision. Do you understand that perfectly. So you can see the superhero that it created for me. I don't want it to be like this, actually. I want it to look like Spider Man. So I would just say, I want the super hero I mean the suits. So I'm just typing my pump as if I'm having a conversation with set. Okay? So, while this is generating, I'm going to quickly take you or show you this next to right here. So this is the Google AI Studio. This is like the central hub for all of Google's AI products, okay? So it's very, very easy to access. You simply type atudioggle.com right here. Or you just search Google AI studio and click the first thing. Right, so this is it. And if you want to generate an image using this AI studio, you simply need to come to this place here, okay? And we're going to be selecting the two point oh flash preview image generation. Okay? You can see it right here. Normally, it's going to be by default, it's going to be in this 2.5. So if you're looking for that two point oh flash preview, you need to select tab display, okay? And then you'll be brought to this section. You simply just tap here and that's right. So we have selected our model. All we have to do right now is just paste this prompt and click Run. Alright. So let's go back. While this is generating, let's go back to this place here. And wait, I think it's done already. That was actually so so fast. I think this is powerful. Let me just say it. I need a full body superho. Let me see what it's going to do for me right here. You can see? So this is amazing. This is exactly what I needed. Something that looks like spider I mean, you can see the suits. Let me come back to this gemina see now. Oh, my. Wow. Okay. It's actually fun to work with all these AI models, okay? So if this one board doesn't work for you, this gem Li 2.54, you can easily just come here and use this two point oh flash preview. I actually think this is amazing. This two point oh, you can see the hot symbol. You can see this hot sign beside it because a lot of people actually find it useful. A lot of people actually use it, okay? So this is amazing. It's came with the white background, so it'll be easy to remove the background. Also, if you want to generate a superhero, you want to generate an image based off another image, like an image of yourself. You can easily just click this Plus button and you can insert a sample media right here. So you simply just click this upload the file. So let me try uploading something right now. So I just added a portrait of myself, I'm going to just add this prompt and use this face for Great. And let's see if it's going to. So these are the ways you can actually create images for your videos, okay, characters. If you want to create a specific character, it's advisable, you know, you create the image first and then feed it into V three. Of course, you can actually just tell V three. Describe what you want, and it's going to actually use Gemini at the back end to, you know, so refine your prompts and create that image that you need. Oh, it's done. But this is not me. This is not me. Okay? So let me let me just the image I inserted, let me insert it again and just do something simple. So you can see what I just did right now. I uploaded my image, okay? And I told it to change my head to a braided hair. I can see what it did. So you can see what I just did right here. I uploaded my image again, and I told it to change my head to a braided hair. I can see the results. It's actually for your bots. You can use. This feature is actually very, very useful if you're creating characters. This could actually be a character in a video that I want to create because it's just your creativity. The limits is your creativity. You can do whatever you want to do using these AI modules, okay? And also, Gemini right here is not bad. It's not bad. Let me just refine this one put fuman. Like man like supero and see if it's going to, you know, type this by anyways. Gemini is smart. So you can see what is generated for me right here, and this is perfect. Let's say, it looks like Spider Man. The whole suits, the Spider logo right here and also added the ePatch quested and the suit is green. So this is perfect. So this works. And if you want to download the image, you simply just click this button right here, that icon right here, and it's going to be downloaded. Al right here, if you want to download this image, you simply come here and you click this download button right here. So you also have this option right here if you click this button, the three dots right here and you click this branch from here, I creates a branch of your current conversation. What mess is that now that it has created, you can see now there's a path here or there's a button here, let me say, there's a text her for me to see the original conversation, which is this AACN supero image, okay? It's the one that contains everything, all these images right here, okay? But now that I've created a branch, I can treat, I can try different things without modifying the original chart. I don't know if you understand that. Me if you create an image and you think you're satisfied with this, but you just want to explore all that sense without hurting the image you created before. I know you might be asking, Okay, if you generate a new image, it's simply just going to generate a new image. I won't change the other one. But you're adding more context to your chart, and just imagine you created a superhero with a cape. Okay, then you now sent another prompt telling the AI to remove the ape from the superhero. Now, the last image that the AI would actually remember is the superhero without a keep. Now, if you now send another prompt to it, let's say you are now so you are not satisfied with the superho that doesn't have a keep, or you want the one that has a keep, okay? Now, you now think that if you send it another prompt, it's going to generate it based on or it's going to make changes based on that superho that has a keep. No, it's wont. Now, if you send proms now that's Google AI, or Gemini should add a sunglass to your superhero. What is going to generate is your superhero without a keep and with a sunglass and with sunglasses rather, okay? But if you want to preserve your original superhero that you are satisfied with, and you just want to make some changes, try out new things, then you can easily just create a branch, okay? So you click this branch from here and it's going to create something like a for you know, in this for, you can create a fork of a project so that you can make changes to that project. So that's the same thing right here. You create a fork and then you can start making changes to it. And when you are done, you are satisfied or you just want to go back to your original image. You simply come here and you click this and you see it will bring you back to the original chart with the unmodified superhero or character, right here. Do you understand. So let this load up. You can see. They can see C branch conversations of you can click this so see the branch that you created earlier, and we're currently in the original charts right now, and this is the last image right here. Okay? So I hope you understand this clearly. Make sure that you practice as you were watching. Okay, it's very, very important. Don't just watch me do stuff here. Pick up your laptop, is AI studio and start trying things. So now I'm going to explain this temperature right here. See, as I hover out my mouse, my cursor over this slider. I see that it show this creativity allowed in risk in responses. So if you reduce this when you drag this slider close that to zero. Okay, it's going to use less creativity. It's going to bring out a more balanced image, less creativity. Okay? But when you increase the slider, you drag the slider up to close that to one or more than one is like a balance. It's in 0-2, okay, two and zero is the midpoint, a balance between creativity and just let me say, normal images. Okay? So when you this increase this temperature and you put it to two, it's going to create unexpected outsputs. I don't know if you understand that. So instead of doing something like this, you might decide to try using let me say gana braids or red brids or something like that. You know, just try crazy things. I think that's the best word, okay? When you increase this temperature, it increases the craziness. The craziness of your outpots, okay? Or when you reduce this, you bring it closer to zero, it's going to reduce the craziness of your outepot. I hope that makes it very, very clear to you. I'm just going to leave this in one, okay? Easily just come here and pipe one. Okay. I was going to bring this here. So that's it. And for this token counts, now, this is very, very important. You can see that we have 2,768 2000 tokens for this chart. Now, what are tokens? Tokens are simply what the AI uses, you know, to generate your image, your images, you know, not just to generate your images, you know, the syntax, characters, and all that, right? So I think they actually counts like, let me see, this one is a token. This one real, a token, or it's just letter by letter. I'm not really sure. But then the toing counts, if you need to keep watching this while you are creating images using Gemini, or using this AI studio. Why? Because if this token count is maxed out, let's say you get to TT use up the whole TTS 2,768 tokens that are right here, the AIs memory is going to reset. Technically not reset, but it's going to start losing context. Okay? So this 32,000, we also call it context window. It means this is the largest amounts of contexts that your charts can hold. So this is very important when you are doing different actuations of your superho. Let's say you're trying to get that perfect image. And you are sending different prompts, and you keep sending trying new tens, trying nutrients. Let's say you've got to your perfect superho but you want to get an image of this same superho probably sitting down or in a car, so you can use it in your video now, you keep using the same chart. You need to keep watching this context window because when you get the end of this context window, you use up the tokens right here, it's going to start forgetting the proms that you sent at the beginning of the chart. Okay? So that means if you're not send new prompts now and you try to reference something at the beginning of the chart, it's not going to remember. Okay? So it's going to keep forgetting things at the beginning of the chart so that it can accommodate the new prompts that you send. I hope that is explained well enough. So you keep watching this. What you can do when you are actually very, very close to the end of this context window. Or if you've actually almost used up these talking counts, what you can do is just tell it to give you a summary of all the proms that you've sent right here, all the proms and outputs, or just say all the proms. And then you can copy it and create a new chart. And to create a new chart, all you need to do is simply to come to this chart icon, this chart here, just click it and it's going to create a new chart. You can see right here, you can see it. I created a new chart, and I can just type what I need to type right here and run it. Okay, you can see the toing counts right here, 15. If I remove everything, you can see that it's zero. Switch back to zero. You can see zero, one, go, go, go, go. Six tokens. So this is 123. This is just five letters or five figures here, and it's counted it as six. Let me add this space and see if it's going to count that as a token too. Okay, so you can see it counts tokens. Characters plus space, every character in your prompt. And this also includes the outpos prompts that Gemini will send you to. It's not just your evos prompts. Every outpos that Gemini will send you to the prompt is going to also count it and add it to the token count right here. Okay? So that's it for this image generation. Make sure you try it out there. Also other tools that you can use like ChargBt. ChargBt actually generates amazing images, but it's quite slow most of the time. Okay? It's also really slow most of the time. So I actually suggest you use Google Gemini or the Google AI studio, which is the Gemini 2.0 flash preview image generation. Okay? Now, there's also one more way that you can create images, which is using imagine. Okay? And I'm going to be showing you this very, very soon later in this course. Oh, I hope you're enjoying this course so far. Please leave a honest review right now. You can always update this data on the course. But if you're enjoying this course right now, please leave a good review and let me know. These reviews help me, help other people to see this course and watch it. Alright. Thank you so much. 10. Asset Integration: So now that we have created our image using Gemini two point oh, what do we do with this image? And how do we incorporate it into our video project? It's very, very easy. So first of all, you download your image. You can see the image that two point oh, Gemini two point oh in AI Studio generated for me, this superhero, it's perfect. Okay? So I've downloaded it. I'm just going to close this up and go back to Google Flow. Okay? So let me just refresh this now. L you just refresh this page. So right here, I want to upload the image of my superhero and make a video with him, okay? So if I click this model now, if I go back to the sentence and I click this model dropdown, you can see right here that VOT first only does takes to video generation. Okay? B does takes to video generation or what just happened. So if you actually want to incorporate images in your video generation, you cannot use this VT first. Okay? So let me just select this Vet first first of all, and then come to this section right here. So if I click this dropdown right here and I select frames to video, okay? I'm going to see these icons right here to upload the first frame or to upload the last frame right here. So I'm just going to click this first one, and I'm going to upload that image that was generated using AI Studio. So you can see the image now and you can easily just crop it by dragging this this is dragging any of the corners, you know, dragging it in or out. I'll be able to crop the image. But this is okay, so I'm just going to click this crop and save. Now, the image has been added as the first frame. So what I'm going to do right here is just to tell it to make this image smile, okay? And note that I'm using this view three first text video, okay? So if I click this, if I hit this bot this submits bolting, you see what's going to happen. That it's automatically switched to the View two fast, and it displayed an error, switching you to a compatible model for this feature. Remember I mentioned earlier that you cannot do image video generation using View three fast. So it has already switched me to the Veto fast. And what I can do right now is just hit this button and see what's going to generate for me. The video has been generated, so let me just play it right now. Oh, well, it did not make it smile. Oh. Okay, so you can see what is generated. And anyways, this is just a test prompt. Okay? So you can still work with this video right here, but I'm going to be exploring that in another lesson. But for now, I just needed you to understand how you can incorporate images into your videos. Now, not something. You can see that it used that image as the first frame. You can see it here. So it's used the image as the first frame. That's because I uploaded the image as the first Okay? So if you want to work with this video and you don't want this to show, you can easily just edit some cap code and cut that part out. Okay? So remember that I mentioned in the previous lesson that there's another way you can actually generate images to use in your videos. And to do that, all you need to do is to simply click this plus icon right here. If you have a Google AI or Tra subscription, can switch to the gregens video, okay? So this gugent video is actually better because it allows you to upload multiple images without having to specify if they should be used for the first frame or the last frame. But I am using the AI Pro plan right now, so I'm going to stick with the frames to video. Okay, so I'm going to click this plus button again and right here, you can see this generate image icon right here. So if I click this, you can see, all I need to do right here is to describe the image that I want to create. Let me just go back here and copy the prompts that I used for this one. Okay? So let me just copy this prompt and I'm going to come back to Google Flow, pass this full body a full body image. Okay, and then I'm going to just hit this button right here so it will create the image. So you can see the different images that's created for me. It's created different images, not just one. And I can just go down and start clicking. You can see. I also added the spider legs at the back that I actually did not want. So if I want to remove this, I can easily just specify in the prompts, okay? By just clicking this use prompts. If I click this use prompt, I can specify now that I don't want I don't want any spider legs on the Super he so I can just send this to you, and let's see what is going to generate right now for me. Well, it seems like I'm going to be here all day if I decide to keep it treating on these prompts, because it's still added these things right here that I don't want. Okay? It's still removable, but I will still need Lou to create more prompts and keep it treating, trying trying until I get the perfect image. In this case, I think the Google two point oh flash preview in the AI studio is so much better and faster. Okay? But once you get the image that you actually need these images are created using imaging, which is like Google's best image generation to right now. If you are satisfied with the image, all you need to do is just click this use image. Okay? If you're not satisfied, you can easily just cancel it. But now let's try and just use this image now. So I'll just click this button right here, and you will see the image appear as the first frame. Okay? That's how easy it is to actually incorporate images to your videos. It's so easy, and I've shown you all the image generation tools that you actually need to generate your images. So get to work, start practicing, and I'll see you in the next lesson. 11. Maintaining Visual Consistency Across Scenes: This lesson, I'm going to be giving you some tips to help you maintain visual consistency across all the things in your video. I know this is a very important topic for everyone. We want to be able to create videos that have consistent characters all true. So I'm going to be giving you some tips that are going to help you to maintain visual consistency. If you don't understand what visual consistency means, what I mean by visual consistency is that if you have a particular character, let's say this boy is the main character in your video. You want to use this boy. You want to make sure that everything where this boy is, the boy looks exactly the same. Okay, so we don't want this boy to change to a man in another scene or to change to a different a boy, a Black guy or, you know, from a black guy, change to a white guy or something like that, who wants to maintain the same person. How do we do that? The first step is to make sure that before you start creating your videos, before you start generating your videos in Google Flow, make sure you've generated reference images first. If you don't know how to generate images by now, you can revisit this section where I explain how you can generate images using Jemi and using the AI studio, okay? So, make sure you generate your characters. Once you've generated your characters, you can easily call them anytime. Okay? So um if you click this jumps now, and also create a different sing, you can easily call your character by selecting this place, uploading or selecting your character from your assets here because you must have uploaded your character here. So let's say we're working with this. These are protagonists in a movie then you can keep selecting this protagonist and, you know, generating scenes with this protagonist in it. If you want to generate a scene in the subway, select, you can see what I did with the Spider Man right there. Generated a scene with Spider Man running in the subway. I want to generate it, you know, running on a skyscraper, we can easily do that, too by selecting the image as the reference image right here. Okay? So you can see how easy it is. Once you have your character, you don't need to start describing this character again. Let's say we don't have this. You would have to start describing a man in a green suit with a black spider on a suit and an ePatch. That's not optimal at all. You know expects Google flow to consistently generate the same carter from your prompt. No, you don't expect that. You can't expect that. So that's why it's very important to have referenced images. Now, if peradventure, you actually generated your sim without a reference image, let's say something like this now. Like we just said a boy, you know, working in the working in the rain in this kind of environment. And then we still want to keep using this boy across our movie. What we have to do right now is to click this icon right here, this same frame as an asset. When you click this, what it simply does is to save this particular fim with the boy drain and all that. It's going to save it as an asset right here. So that you can easily call the Sasa. You can easily upload this asset, you know, select it and make use of any of these things on any item, let's say, any item or any character or anything in this particular frame. Okay? That's how you maintain visual consistency. If you're creating long videos, it's very, very important that you maintain visual consist so that your movie actually feels like a removie and not just like, cut and join. Now, what if you're actually generating videos one by one? Because there are times whereby you actually. Let's say you are using View three, for instance, you can't use View three in this same builder, obviously. And let's say you really want to use View three first to generate the audio. So in this case, sort of using this same builder, you're going to be generating the videos one by one. You don't even need to come here. So you have your scene, you have your frame, your reference frame, and then you just plug it in. Okay, once you go to the normal Google flow, and then you try to generate an image, you come to the frames video option, okay, or to the ngogens video option. And then you upload your frame and you generate your video. Now, after generating your video, if you want to generate a new video to Axis, because now we're not using the same builder. We're going to us on an external editing software like Capcot. We're going to be doing this later in this course. Okay? So once you have your first s, you also generate another scene that still has the same person in it, you have to make sure that you have that frame, okay? You have the particular frame saved as an asset. Now, let's go back to the normal Google flow. But before I exit this same builder, let me first of all show you how to download this. So you can select any frame anything here, it doesn't matter. And just click this download button right here. I just hover your car over this particular this thing right here or this canvas and hit this button, is first of all, going to export the video. So that's going to take some time. So when it's done exporting it, you're going to see a pop up right here, a little pop up. Showing that it's been downloaded, it's been exported already, and you can easily download it. Now, before you exit this sing builder, make sure you download your video. Okay, assets now, I don't really see a way to come back to this sing builder if you exit it because if you go back to the flow home page, all you are going to see is just the individual scenes. Okay, the individual sin. You're not going to be seeing everything that you created right here. I think you're still going to need to work on this so that we can easily just access all the scenes that we've put together, okay, and work on them again. But for now, there's no way to do that. So make sure you download your scene before exiting this place. Okay? And lastly, I don't know if I'm going to talk about this later on. So let me just do that now. You can see this arrange icon right here. If you click it, you're going to see the different scenes that are in this particular video. You can see that you can easily just drag it, okay? You can easily drag it to swap it. So if you don't want this one to be the first scene again, you want the one with the Spider Man to be the first scene. So you just hover your cursor over this scene that you left click and hold and drag it. Okay, whenever you see this line right here, you can easily just let's go, and it's going to appear in the middle in between the scene and the scene, okay? So if I bring it here and I let's go, it's going to drop the video before the second film and after And if I want to bring it to the front, you can see it right here. If I drop it here, it's going to become the first frame or the first scene right here. So that's how easy it is. And you can easily remove anything that you don't want to here. You can see this minus icon. So let's say Google flow generated a scene that you don't like at all. You can easily just come here and hit the button hits this. Deletes it like that as easy as that. So that's it. You can just hit this dumb button there and it's going to bring you back here. So I'm very sure you now understand fully how this same builder works. Okay let's say the delete button is not obvious right here. If you want to delete, like I just showed you, click this range and come here and you just hit this minus icon right here. If you hit this minus icon, it's going to remove it. Okay? It's going to delete this. And I don't think there's any I I use contro us, nothing is going to come back. So you see a lot of work that needs to be done on this Google flow, and I'm pretty sure that as time goes on, it's going to advance. Okay, so then just hit this do button now and go back to this flow to the flow homepage. So we're back to the flow homepage, and you can see right here, there's no option for me to select a same Builder, to go back to the scene that I created. So if I just click this now, this particular projects that I've been working on, you're going to see the different scenes that I created in the process. You see the individual scenes that I created. You see? And the prompts to the proms that I used see everything here. So what I was there to show you is that if you generate an individual scene, let's say you are just generating individual cliffs and then you are exporting them to cap codes and joining them together. If you want to generate a new sin while still maintaining consistency, character consistency, and environment consistency you can do is come here, you know, just come here to the scene and you click this ad to sing. Once you click this ad to sing, it's going to open the Sing Builder, and then you can easily just select the jump two and then select a fs film from this B right here, there's no option for me to directly add this as an asset. So I need to open the sing Builder. I need to add this to the sing Builder and then use the jump feature and it's going to select last parts as an asset for me. It's going to select the last part as the first film for my next video. So it's as easy as that. That's how you maintain visual consistency. Lastly, what if you actually want to use View three? Okay, you want to use Vio three fast and you want to generate audio. You know, if we add this video to the sibler, we're only going to be able to generate a video using Vo two fast, and that will come without audio. But if you want to use this View three, how do you do that? The bad news right here is that you simply can't simply can't because View three right now doesn't support adding of frames or ingredients to video. Okay? So that's why we are always going to be using Di VwTFas. But what you can actually do is to describe this scene. You can take a screenshot of this scene, okay? And then upload it to Google Gemini and tell it to describe it. Okay, so once you just click this icon and you maximize it to full screen, you can take a screenshot. If you're using the Windows PC, using the DO PC, you can use the FN Windows and print screen or you can even use your snapping to anyone. Okay, make sure the videos are the highest quality. You can even export this first of all at one ETP, then take your screenshots. Once you take your screenshots. You can take it to Google Gemini, and then you upload it you upload the image. Once you upload the image, you tell Google Gemini to describe this image for you you tell us that you want to use it in V three, okay? So let me tell it to describe the image for you. Now, once it has accurately described the image for you, you can copy the description. Come here, select Veo three, and then you simply add whatever you want to add to it. It might not be the same thing as this one, but I'm pretty sure it's going to be very, very close, okay? So that's it, and I'm going to see you in the next lesson. Keep creating. 12. Camera Controls: I wonder why some AI generated videos feel flat while others feel like they were shot by a Hollywood cinematographer. The sequent is camera movement. In this lesson, I'll show you how to take full control of your virtual camera inside Google Flu, from smooth spans and dramatic zooms to aerial shots and dynamic angles. Because in filmmaking, even AI filmmaking, where you place your camera changes everything. They make your shots look intentional, cinematic, and alive. So let's get started. There's a biting feature here in Google Flow that allows you to select camera movements, okay? So before I work with this, I'm going to use these prompts that I used for this desert boy, okay? These same prompts I use for this desert boy, let me first of all, just scroll down a bit and reuse this, okay? I'm going to use this and let me see. So you can see the camera movement that I added to my prompt right here. It's still the same prompts we'll be using from the beginning. But I just summarize this so I made the song shorter, okay? So you can see right here. I told View three to make the camera shots begin as a medium view from behind and slowly rise up to become a full aerial shot from the top. In other words, it's just an aerial shots that looks like it's a shot from a drone, okay? So I'm going to clear this off. I'm going to remove this now. So if I send these prompts view three like this, it's going to create a flat video, a video without any camera movement. And that's why I added the camera movements by myself to these prompts because camera movements are very important in creating cinematic videos. But now that I've removed it, what else can I do to create or create that same camera movement? I simply come here, click this drop down and come to frames to videos. Okay, frames to video, and then we are here now. So look at this icon right here. If you click this icon, you are going to be seeing different camera. Let me just zoom this in. So that you can see everything. Okay? So you can see different camera movements right here. We have the dolly in the dolly outs, Jeep down, jib up orbits left, orbits right, and different camera movements right here. Okay? So if you want to preview what the camera movements would look like in your video, you can simply just hover your curso over it. So you can see what this looks like. This jeep down. You can see this dolly out. See what this dolly house looks like. You can see you hover it over the span to the right. You can see the way the ship in this, I think is a ship. It should be a ship. In this video is actually panning to the right. Okay? Now, if I hover over the jeep up, look at this. Do you notice any similarity with the video that we created earlier? You can say that it's the same thing. So that camera movement that I described in my proms day, you know, the medium shots from the back then rising up to an aerial shot is actually named Jib up. Okay? It's called Jib up camera movement. L can easily just select it right here instead of adding it to my proms. Okay? And that's exactly what I'm going to do right now. So I have slighted Jeep up, so I'm going to hit this button right now, and let's see what it's going to create for me. So you can see what just happened right now. I will switch back to the View two fast because I'm using these flames to video, and every feature under these flames to video is not compatible with the View three first, okay? I mentioned this earlier because you don't even need to worry. Once you select the frames to video and you use any feature, the first frame, last frame or the camera movements, it's automatically switch to video too fast. So these are the videos that were generated for me, and I'm going to play them right now so that you can see the camera movements that I selected. So, watch this. You can see the weights jibbing downwards, you can see. Okay? And then let me just play this second one too. You can see the same camera movements, where it's going downwards. Okay? So if you notice there's a similarity between the camera movements in this video and the camera movements in the videos that I have been generating earlier, we actually specified how the camera should move. Even though these ones are not a smooOdi not even better than the one I created earlier, but this still works. The uniting Odia is just the distance. The guy is just too close to the camera, but it's still fine. Okay, so you've seen the camera movements, and also one more thing. Asat now. The only model that allows you to use all the features under frames to video is De View Too Fast, okay? Only De View Too fast allows you to select the first frame, the last frame, and the camera movement, okay? As at the time of this recording. So I don't know if it's going to change later on. Whenever you're watching this video, maybe a year from now or six months from now or even three months from now because the way that switch updates are being applied to this Video three and every other AI two out there is amazing. Like, it's really amazing. It's fast. So that is for camera movements. Now, if you've been paying close attention to all the videos that have been generating so far, you'll notice that this video doesn't follow this prompt. This part of the prompts where I specified that the video should be a hyperrealistic video. If you notice this video is something like a gibbly art style. Remember the trend where everybody, including the mosque, were creating gibbly arts variations of their photos. You know, this actually looks like it. And I don't know. I think it's probably because of the trading data. You know, the period when people were generating so much Gibby style images, I think it somehow messed with this, but I think it's going to be fine later on. Oh, I encourage you to start practicing right now. Just write any prompt, select a particular camera movement. It could be the panel to the right, panning to the left, the jib up or Jip down, and then send it to view too fast, see what it generates for you. You can also try experimenting to see the difference between using the jeep down and specifically telling it to use an aerial shot. If you don't understand what I'm talking about, you can easily just check previous videos and check the prom that are used for the previous videos that are generated, okay? The parts where I said, it should be a medium shot rising up and becoming an aerial shot. So you can experiment with that, you know, and just notice the difference, okay? So just play around and I'm going to see you in the next lesson. 13. The Scenebuilder in Google Flow: This lesson, I'm going to be introducing you to the same Builder in Google Flow. And this might be the most important lesson in this course, so I need it to pay close attention. It's very, very important to understand this lesson. Okay? So let's get started. What exactly is the same Builder? The same Builder is the whole point of Google Flow, because you can easily generate videos in Gemini. There's also a website for V three, where you can go there and Genens videos, okay? But these videos are only 8 seconds long. Now, what if you actually wanted to create a video that was longer than 8 seconds? How would you do it? Now, that's why there's Google flow, okay? So in this Google flow, we have a feature called the sing Builder, and this sing Builder allows you to extend your clips, okay? That allows you to extend your clips, make them longer, add different scenes, you know, new scenes, your clips and essentially create a movie if you want to create a movie, okay? So let's start with this little scene that I created, okay? Just a scene of a guy of a little boy walking in a stormy atmosphere. Look at it. So you can see what it looks like. So if I hover my Coso over this particular scene right here, you can see here at the top left answer, you can see this art to sin option here. So just click this, click this and see what happens. See that we've been redirected to the s Builder in Google Flow, and this is exactly what it looks like. It looks like an editing software. You can see the timeline right here. We have different options. We have this plus icon. We have a particular the particular scene we're working on right here. We have the option to download. We have this rand. And yeah, we have the normal interface right here, it takes video view too fast and all that. Okay? First of all, you can cope this scene, okay? By just putting your co, as either of those sides can put it here and copy it backwards just, you know, to remove the sides that you don't want, okay? And I can just extend this back. And make it snowmel. You see? So if I want to extend this clip, I want to add something new to this particular clip and make it longer. This is what I'm going to do. I'm going to click this button here, this plus icon right here. And here, there are two options, okay? So I need you to pay close attention right now. There are two options right here. We have the extend, and then we have the jump too. Now, what's the difference between these two options? The extend simply extends. Yeah, I simply extends or makes a particular clip. The particular clip that you selected, it makes it longer, okay, by adding new sense something new to it. So you use this option when you are working in a particular sent you just want to work with the sin. You don't want anything to really change. You don't really want the environment details to change. You just probably want to add something new to it. You know, you're not directly changing to a different senataly. I don't know if you understand that, but that's what this extends for. Now, let me show you what I mean. So if I click this extend option right here, you can see what shows, what should happen next. And if you look at this now, I see the way this fades out. That means this whole part right here is going to be another sin that blends with this first sin. Understand. So it goes from here and it smoothly transitions from one scene to another scene in the same context, okay? So if you click this extend, you have limited options. You don't have the option to add the four frame or last way, basically because you are using this particular frame right here. So what I'm going to do right now is just to type Okay, see what I just typed right here. The boy approaches a gorilla sitting beside a tree. So simply, what I want right now is for this guy to keep walking, and then a gorilla just appears out of the blues or out of the blacks, in this case, and, you know, the boy just walks towards the gorilla. So that's what I want Google flow to do right here for me. Now, before I hit this button, you need to understand that only viewed too fast works for this. Okay, everything in this seeing via, you know, the extension of your sing and the jumping too, only works with VeledToFast. Everything else, you know, Vo two quality won't work with this. If you try to select any of those models, you'll be automatically switched back to view too fast. So if you're planning to add an audio right here, maybe to make the boy talk to the gorilla well, bad news is that you can't do that. You can only work with a video that does not have any sound. So I'm just going to hit this busy right now. And you can see that it started generating. So let's wait for this to be done. So the prompts I sent earlier got flagged. I don't know why, but I just changed it to a lonely puppy. Maybe the gorilla was too violent for Google Flow. I don't know, but I think puppies are cute, so this should work. What I'm going to do right now is just to copy this. Let me just cut this out, and then I'm going to extend this clip. Okay? I'm going to paste this right here and send it again. So it has been generated already. You can see the duration has almost doubled. It's went from 8 seconds to 14 seconds. Okay, so let's split this now and see what it looks like. So you can see what just happened. So you can see exactly where this first scene transitions into the second scene. It's somewhere here. Okay, so I can just drag this slide right here and you can see where it transitions into the second scene that has the tree and the poppy. You can see the puppy right here. Let me expand. Let me make this full screen so that you can see it very well. So you can see now, if you notice the environments in this first sin and this second sin are the same thing. You can look at the mountain here. You can look at this here and you can look at the same mountain right here. So you can see that it maintains the same environments from the first scene, okay? Because of these mountains, you can see how identical they look. And also the horizon, the horizon is also a little bit identical. The environments, the lighting, the rain can see it's identical. And that's the whole purpose of the extent feature in Google Flow. Okay? So let's move on to the jump two feature, right? So if you come here and we select this jump two, now, you need to take note of the clip you have selected before you select any of these options. See, some people make this mistake. When they want to extend the particular clip, they don't check to see the particular clip that is selected. I've seen a video where someone selected this particular, the first clip, and you wanted to actually extend the second clip. Like I said, we have not seen yet. It's just dark. There's no Tmn, there's no preview right here, but we know that there's a particular clip here, okay? You and I know because we've been creating this. So people make the mistake of selecting this and then pressing this and extending it. Okay, now, if you extend this, it's going to extend this clip right here. It's not going to extend the last clip. Okay, so you need to take note of this. Make sure that you select the last clip if it's what you want to extend. Okay? So now, I'm going to click this, make sure this is selected. And now I want to show you what this jump to is useful and how it works. Okay? This one is a little bit tricky. This jump to allows you to change scenes. Okay? But when you select this jump to, see what happens right here. It automatically selected the first frame. And this first frame is the last frame from this scene, right, say, the last scene. Okay, the scene I selected, it's picked a frame, the last frame in this particular scene, and used it as the first frame of this new scene that I'm about to jump to. I mean, why does this jump to feature actually exist when it does just the same thing that the extend feature if we select this last fim from this particular sin as a force frame in the sine, it's simply going to do a smooth transition from this last scene into the sine and it will be the same environment. You see, that's the problem, because some people struggle with trying to jump into a new sine, not knowing that they have to remove this fs im. Okay? Now, if you remove this false film, you can't walk, you won't be able to send any sin. So let me just say, then another dog. Well let me say yeah, if I type in then another dog appears out of the blues. Another dog walks to wards them. Now, I'm not going to send these proms because I don't really want to waste time here. So if I hadn't removed that force sin, okay, if I hadn't removed that force frame, what would have done to add that dog. In this same environment, you see the dog walk towards the boy and this was that dog right here. But let's say we now want to generate something like this boy want this boy to be in a different city or not just create an entirely different scene. Okay? How do we do that? We need to add a false frame right here, okay? So if I select this now, you can see all the assets that I have here. You can upload the new assets if you want. You can generate an image if you want, okay? And you can select. Also, you can if I come here and I come to this part here, you see, I can easily just save this frame as an asset. Okay, you can see the option. This particular frame, I can just click this plus icon, and it's going to save it as an asset right here, okay? And then I can easily use this asset to generate another scene. Okay, so this is very, very useful if you want to maintain consistency in your movie. So make sure you take note of this. It's very, very, very important. You can see that it has added it has added the frame already as an asset for me. So I can decide to select this later on in the movie that I'm creating, okay? I can decide to select this and do something. Okay, let's say, I actually wanted to talk about this in another lesson. So I think I will talk about this in another lesson. For now, let's focus on this jump to option, okay? So I want to jump to a scene of this particular character, this superhero running on a subway or running in a subway, okay? So what I'm going to do now is to select this as my first frame, okay? So I'm going to select this ask my first frame. Take note. Okay, first of all, I did not select anything here. Let me just You should have only one option. So jump to. And yeah, I have only one option here. So once this loads, I'm simply going to cancel it and add this as my reference frame, okay? So at this point, it's not just a first frame. It's our reference frame to create this scene. Okay, so what I'm going to type right here is the character So you can see the proms that I typed here. You can see this is what I want it to jump to. From this scene, I want it to jump into a different scene. So make sure you select a different phos frame from what's flow is going to automatically select for you. Okay? So I'm just going to hit this button and see what's going to generate for me. All right, so the scene has been generated. Let's play this to see what it's going to look like. So you can see what just happened. It switched from this scene right here, okay? So this Spider Man scene. You can see how it happened right now. So if you had typed in this prompt, you know, for it to generate a scene of Spider Man, well, let me show you what I let me show you what I'm talking about. No need to just talk about it. Let me just go to this jumps now and I'm not going to remove the last thing the first frame that Google Flu is going to add automatically for me. So I'm just going to say an ARAchN An AACN superhero. So see the prompt I type right here. It's practically the same prompt I used right here, and I selected this, this film of this, you know, this boy and the puppy. Okay, so let me just sing this right now. Okay, so it has generated the scene for me. So let me just play this now. So this is the scene of the Spider Man and the subway, the first frame. And now you can see this. Okay? So you can see what just happened right now. It's basically transitioned smoothly from this particular scene with this boy. Now, it's transition into a subway, right? But you can see now this rain is there. Why is rain falling in the subway? And why is this boy white here instead of Spider Man? The answer is simple because the first frame that was selected while I was generating this particular new scene was the last frame from this particular scene. I explained this before. So it's used the last fran from this particular scene. And in the last frame, we can see rain here, we can see rain falling in the background. You can also see this boy white head. So he picked the boy and use the boy as Spider Man, and it also put this ring in the subway. So that's the difference, and that's what you need to watch out for. Make sure that the new scene you are generating has a frame, a reference frame. Okay? So you need to pick a reference frame so that you can accurately generate a different scene from what you have in your video. Okay? So if you want to generate a scene that still works with the same character in this video, then you don't need to change any scene. So you can just automatically can just use what Google Flow automatically generates for you. Okay? So it's as simple as that, and that's the difference between the jump two and the extend features in Google Flow. So I encourage you right now to go try it out. Use the jump two feature and use the Extend feature and see what happens in both cases. Okay? Also, don't forget that it only works with ViewTFast at this moment, okay? At the moment of this recording at the time of this recording, it only works with View too fast. In the future, it might work with View three. I'm pretty sure it to work with V three, okay. But for now, it doesn't. So if you're watching this course, when it works with View three, you can can just select V two and add audio to it. All right. So that's it for this lesson, and I'm going to see you in the next lesson. 14. Asset Management: Organizing and Managing Creative Elements: Now, let's talk about assets management. What are assets in this context? Assets are everything you create and use in your project. They include images from imaging, the videos from view, the voice overs from AI Studio, the music, ambient sounds, and evil scripts or proms documents. Okay, everything, all of them are called assets. Once you start generating a lot of visuals and audio, it's easy to get lost in the mess of downloads. You forget where you put a particular image, where you put a particular video, you forget where you put a particular proms document. Oh, this will slow you down. It reduces your efficiency. And you definitely do not want this, especially if you're creating multiple projects. You will get confused. So in this lesson, I'm going to be showing you how to organize your assets. So I'm just going to show you how I organize mine, okay? So this is my file manager, as you can see right normally, this is how I organize my assays for video editing. I have a note edited folder for videos I've not edited. Then it moves on to the edited folders, which are the videos that are edits from Capcod and then the compressed handbrake folder, which are the videos that I have compressed already, and I'm ready to upload them to whatever applas for my wish upload them too, okay? And now I have this new folder right here, this AI Gen video. So you can name this anything you want to name it, okay? The pot is under the video section. So you can also save it on your desktop. You can add this to your desktop so you can easily find it, right? So I have this folder right here, and if I open this up now, you can see the different folders that I have in here. I have the audio folder, the editing videos, or the edited videos, rather, I meant to type edited videos, okay? Edited videos. I have the images, reusables, sins scripts and prompts, okay? So the scripts and prompts folder is where I save my prompts and scripts, okay? You might be thinking that the prompts that you use on Gemini, I mean, they are saved online, right? They are saved on Gemini. You can easily find them there. But what if you actually go online? You go on the Internet. This has happened to me a couple of times, by the way. You go on the Internet and then you see a particular prompts format that you like see or someone just shares his prompts, or probably you go to FloatTV and you see a particular prompts that you would love to use, how would you save that? If you leave the Internet, so if you leave float TV, you might not find that prompts again. So it's always advisable. You copy the prompts. You can save it as a text Document using your notebod or Microsoft Word or Google Doc, and then export it as a PDF or TextFile TX Cfle. Then you save it in this folder, so you can easily access it later on. Okay? Then this SNS folder is where I save every scene that I download from Google Flu. Whenever I generate any scene, I come here, you know, I save it. If you download my downloaded folder, I simply move it to this sen spot so I can easily find them later, right? And then this reusables folder is just the same as the SNS folder. The only difference is that, yeah, I'm going to be saving scenes that contain environment shorts. Like a scene that doesn't contain a particular character. Let me say, a scene of a mountain, an area short of mountains or the woods or animals, something like that. Bs, essentially, which are videos that I can use in different videos. Do you understand me? So if I generate a Bo footage in Google Flow, I can always reuse that BO footage in different videos. So instead of saving it in this sins folder, where it's called gets lost in the midst of other footage that I don't plan on reusing. I would rather just save it in these viewable, okay? So whenever I need a BO footage, I can easily just come here or something else that I know I can reuse, okay, I can easily just come here and pick it, okay? So for this images folder, this for the characters that I generate in Gemini or AI studio, okay? Wherever I generate a character, I simply download it and I move it to the images folder, so I can easily find them later. Then these edited videos is for the videos that I've edited in cap codes. Like, I've edited this completely, you know, added the sound and all that. And then once I've exported them, I make sure they are saved in this folder so I can easily access them later. Now this audio folder is for the voiceovers that I generate. Don't worry. Later in this course, I'm going to be teaching you how you can generate voice overs for your videos, okay? So it's also for the copyrights, free music that I download online, for the ambient sounds that I generate, you know, audio files generally, I'm going to have them saved right here. So this is basically how I organize my assets, right? So I know that scene. If you come to this Google flow right now, and if you check if I open my download history, it's because it's showing dited because I moved it to that folder, but it's actually not dited. This video right now that I downloaded from Google flow, you can see the way it's saved. Say the naming convention, June 3 0952 31 seconds, the days and this This is not a proper or optimal way of naming your media files, okay? Look at this one, too. Look at this Gemini generated image here. You can see this no when you download your assets like this, you can look at it, if I come to this sins folder, look at the name here. Make sure you come here and you edit it. So this video is the video about this boy this boy walking in the ring. So I can easily just save this as boy in the ring. Let me rename this. Boy in the in the ring that I can put in one. Good. Now, naming your file like this ensures that whenever you need it, you can easily find it by using the search by. Okay? Even just scrolling through, if you have your video files just named like that, you know, not named properly, you won't be you'll have to open every single video to watch it as it's the right where you need. But when you name it like this, you'll be able to easily find whatever assets you need. The same thing goes for the voice overs, for images and every other thing, okay? So make sure that you name all your files properly. Okay? It's very, very important. And also add tags. Now, if you go back here and we come to the reusables, there's nothing here obviously. But if I had two different videos, let me say, two different mounting videos, one was in the day and another one in the night. You can easily just let me come back here so that I can show you what I mean. You can easily just renm it and put something like night, okay, so that you can easily find the footage or the Bo that was shot in the night. Okay, the nighttime Bro. And for the daytime one, you can easily just put today or any other tag, add tags. Okay, so it's will help you to find your assets whenever you need them. So you need to think like a content studio. Even if you are just one person, having your assets organized makes you faster, more professional, and even more creative. So this is a project for you to do right now. Go to your file manager, create a folder like the one I created for my AI generated videos and structure everything the same way as structured mine, okay, create the necessary folders for audio for videos, for prompt easy, creates them right now, okay? Don't procrastins, create them right now, and I'm going to see you in the next lesson. 15. Natural Language Processing: Refining Prompts for Desired Outcomes: This lesson, we're going to talk about prompts and why they matter more than you might think. If you've ever generated a video using V three and thought, Hmm, this isn't what I imagined. You are not alone. The truth is the results you get from AI tools like V three are only as good as the prompts you feed them. That's where natural language processing comes in or NOP for shorts. NLP simply means how AI understands and interprets the way we speak or write. And that's a big deal in AI filmmaking because vague prompts yield vague results, but refined cinematic prompts yield amazing results. For example, take a look at this prompt. A man in a city at night. This is boring, random, and vague. So instead, you can type something like a tired man in a trench coat works under fliquion signs in a ini Tokyo alley. And these prompts right here is much better. Now I'm going to show you how to refine a prompt using Gemini. Let's start with this basic prompt. A woman finds a magical book in the forest. Now to refine these prompts, we are simply going to add this. We write this as a cinematic video prompt for an AI to add emotional to visual details and atmosphere. Okay, as simple as that. And don't worry. Every prompts template I'm going to be using in this course are included in the prompt pack that you're going to find in the resources section, okay? So just go check it out and you'll be able to copy this. This one will be there and every other prompt templates I've used, okay? So this is it, right now, I'm just going to hit this submit button and see what it's going to generate for me. It has generated a few opsits for me and you can see this. You can see the different prompts that generated, see? Look at this. Look at this. Look at this. You can see all the prompts that I generated. I made them very, very detailed. And these prompts might be a third bit too long for view three. So I can simply just tell it, the prompts make the prompts brief for use in view So you can see a young woman discovers a magical glooming book in an ancient sun dappled forest. The book rests in the roots of a giant's oak. It's cover gloom with golden rings, cinematic cereal gold rays, all and wonder fantasy eight K. So you can see that it generated different prompts for different moods, dread and horror, quiet mystery, and awe and wonder. So if you have a particular mood or emotion in your head that you want Gemini or that you want V to create your video with. So instead of telling it to add emotional tone, you can just specify the particular emotional tone that you want it to use, whether the quiet mystery or something else. You can add it right here. I see the difference between these vague prompts that I put here. A woman finds a magical book in the forest, and the prompts right here a young woman, this particular prompt right here or any of these ones, you can see that there is a clear difference between these prompts. So it's a wise thing to refine your prompts using Gemini before using it in Google F. Also, you can use Char GPT, you can use Deep Sk. You can use any other AI too that you like. The reason I'm using this gemini is because it's obvious this is the pro version, you know, and it's going to finish in a month. So I just want to, like, Mo kit, you fill me. Okay? So do you remember the proms template that I used E in the lesson? This one, let me piece it now. Look at it. This particular one that I used early in the lesson. Now, if you want to use a prompt template, if you found a prompt templates online or a prompt templates from the prompt park that I'm going to include in the resources section of this case. This is what you can do. First of all, put enclose this in quotes. Then I'm going to copy the prompt that I use for this lady, okay? A woman finds, let me just copy this. Right. Then I'm going to tell eminiRfine this prompt. This prompt. Following by following the templates above send. So let's see what's going to generate for us. So you can see that as refined the prompts using the prompts templates that I provided it, the time of the day, the color and mood, the camera lens and position, right, the character's description, and the static shorts and environment details. So this is a beautiful, well detailed prompt. Now, you need to understand that there is a thin line separating too much description and too little description. So you need to find the balance. If you give you three lots or large forms, like too much words. You said that when it generated the first one, I told it to shorten it. So if you give you so much large forms, rather, it's not it's my notes. It's most likely not use everything or generates everything that I specify in the prompts. Why? Your video is just going to be 8 seconds long. Don't forget. So you need to make sure your prompts are brief. Before you send your prompt to view three, ask yourself this question. With everything I've described in this prompts logically fits into an 8 seconds video. If it's not going to fit in, then you can manually remove redundance details from the prompts, or you simply just tell Gemini to summarize or shorten the prompts for you. Okay, you can even tell it that you want to use it in a video that is only going to be 8 seconds long or a video that is going to be generated using View three, and it's going to shorten the prompts for you. So that's how you refine your prompts. You know that creates amazing videos using View three. Now, if you noticed, I've always been adding cinematic cinematic cinematic cinematic. You don't always help to add cinematic. Okay? There are different kinds of videos that can be created. There are cartoon style, anime styles. So if you want a specific kind of video, if you don't want a cinematic video, you want an anime style video or a cartoon style video, you can specify it right here. It doesn't always help to be cinematic. Do you understand me? So to conclude this lesson, you need to understand what actually makes a good prompt. Your prompts must contain character, action, setting, mood, and style or detail. So the character is simply the subject or protagonist of your video, depressing, let's say, if you're creating a video about a woman in a forest, basically, the woman is the character, then the action is what the woman is going to do in the video. Then for the setting, this includes the visuals and the lighting, the environments, where she is, how everything is going to be like. Then the mood, it could be mysterious. It could be horror. It could be Sci Fi ITC, and the style or detail can include the cinematic videos, cartoon videos, anime, and Was not. One last thing you can always add is the camera angle. Okay? Google Flow actually has built in camera movements that you can select while creating your videos. But a better way to do it is even to include your camera movement, camera angle in your prompts. If you remember the video we generated earlier in the course, the word of the boy walking in the desert where I said the camera movement should be like a medium shot moving up to an aerial shot. Remember, I also tried to replicate the camera movement by using the Jep up but in camera movement in Google Flow. And the video where I actually typed in the camera movement into the prompt actually showed the camera movements better than the video where I used the built in Jep up camera movement. So this is something you should consider while creating your videos. You can always experiment by using the camera movement, the butt in camera movements. And if you're not satisfied with the results, you can always add the camera movement manually into your prompt. So here's your takeaway. If you want nematic results from view, start thinking like a director and let Gemini be your writing assistant. Just describe what you want, then ask Gemini to refine it. Use emotion, use visuals, use stone, and you start creating video prompts that actually feel like scenes from a movie. In the next lesson, we'll take it a step further. I'm going to be showing you how you can build out full scripts using Gemini for your short films. See you in the next lesson. 16. Script Development: This lesson, I want us to talk about something that is going to shape the way you tell your stories with AI, and that's script development. As someone who generates videos using AI, you're not just writing poems. You're also directing how your story will be told. In AI filmmaking, there are three main types of scripts that you can generate with Gemini or Char gibt or yourself, and each one of them creates a completely different kind of video. To show you what I mean, we are going to use the same story line 23 different ways. So here's the story. A young boy is lost in the desert. He works for days under the scorching sun, exhausted and noon. Just when he's about to give up, he meets an old man who gives him water and leads him to safety. Now, let's look at the three ways you can turn this into a film. The first is the narrative voice over a script. This one is probably the most popular and it's very simple but powerful. It's just one voice a narrator, telling the story while the visuals play out on the screen. You're not going to hear your characters speak. Just going to be hearing a storyteller guiding you through the emotions. So this is perfect for poetic videos, documentaries, YouTube style edits, where the voice over carries the emotion, YouTube shots, YouTube automation channels. So this type of scripts has a very wide application, and it's probably the only type of script that's going to 100% work for you right now. Second one is the dialogue script. In this kind of scripts, you give the characters voices. You'll see the boy and the old man speaking. My be just a few lines, but the interaction is going to drive the story forward. So it's going to feel like a short spin, personal, emotional, and it actually gives you the chance to work with multi speaker voiceovers or character driven story telling. Okay? In essence, this kind of voiceover is going to include dialogue. So the man and the boy are going to talk. Okay? They're going to talk. They're going to interact. They're going to see things to each other in this kind of script. Do you see the difference between this one and the narrative voice over script, which I explained earlier. The narrative voiceover script doesn't include dialogue, but the dialogue script includes dialog, okay? And then the last one is the visual only script. This one includes no words, no voiceover, just movements, sound design, and visuals. Think of it like a music video or an art hm. Every emotion is shown not spoken. You use ambient sounds, cinematic music, and visual storytelling to carry the entire meaning. No dialogue necessary, no voice over necessary. So right now, I'm going to go ahead to generate these three different kind of scripts for this particular story line so that you can see exactly how to generate scripts for your videos. Okay? Also, don't worry at all. Like I said earlier, I'm going to be including all the prompts that I'm going to use in this course, including the ones that I'm going to use in this lesson right now in the prompts pack. They are going to be the prompts pack. All you have to do is to download it and you can easily copy it and tweak it however you want, right? So let me just pace this right now. You can see the prompts that I created for the narrative, voiceover script. You can write a short cinematic script with voiceover narrative only no dialogue between characters. Then I added the story the story line right here, told it to break it into three different scenes. We're going to be using this one to generate the voiceover for a video. And this one, what happens visually in the scene is simply what we're going to use as a guide to generate our videos in View three. And they suggested music and ambient sounds. We can easily just add the music ourself and the ambient sounds ourselves using calcd. I'm going to talk about this later in the lesson. Right then lastly, I added it to emotional poetic survival them with a sense of hope at the end. You can see the keyword right here. I told it with voiceover narration only, no dialogue between characters. This is the keyword right here. Okay? So, make sure you add this when you're trying to generate a narrative voiceover script. Alright, make sure you add it. And now I'm just going to hit this submit button and see what is going to generate for me. So you can see it has generated it already for me. And for the scene one, you can see the visuals that it's generated. Okay? This visuals is what I'm going to follow. So generates every scene in the video, okay, every eight second clip for the video, right? So I'm going to be following this visual, this visuals part. And then for the voiceover, you can see here. Okay? So when I want to generate the voiceover, for this particular scene, I'm simply going to copy this and I'm going to generate the voiceover. To make it say everything here Vb a thing, okay? I'm going to say everything here. It's not a prompt. It's basically a script. All right? It's words I want the voiceover generator to generate for me. I'm also going to talk about this in another lesson in this course, okay? So the last part is the music and sound. You can see a sparse, haunting melody, a single cello, or you can see it. And then the ambient sound, there's no dialog of any sort in any of the scenes. Okay. So this is what the narrative voice over script looks like, right? So now, I'm going to go ahead to generate a dialogue script, and this is the prompt I'm going to be using. You can see this. Write a three scene short film script that includes the dialogue between two characters, the young boy and the old man. You can see. Then for each scene, include, blah, blah, blah, blah, blah, blah. So I'm just going to hit the submit both scene right now and wait to see what emina is going to create for me. Alright, so it has generated the script for me, and you can see now. So in this one, here, the sun is blot out. These are the visuals, okay? And then you can see now that it has added the dialogue, you see? I didn't mean to. I just wanted to see the lizard. Where did they go? Where did everyone go? It's actually talking to himself while it's here, okay? And then if I scroll down to where he actually met the old man, you can see here that the old man is talking. You see? You can see that it generated the visuals, also generated the sounds, and most importantly, he generated the dialog, and this is why this is called a dialog script. So I'm going to talk about which one you should actually choose later on in this same lesson. So just hang on while I explain the last type of script. All right? I want to generate a visual only style script for the same storyline. So this is the prompt I'm going to be using, you can see here writes a visual only cinematic script, no voiceover and no character dialog. This is the key word. I'm not even going to say keyword, because this is more than a word, but you get the point. This is the important part of your prompt where you're going to be sending to Jeremy I, okay? But basically telling it to create a visual only script, a script that is going to just show that's going to help me generate a video that's going to be all visuals, no voiceover, no dialogue. The story should be told entirely through visuals and sound design. I had to mention it again because Gemini can be so stubborn. So right now, I'm just going to hit this submit spots in right here and I'm going to see the kind of script that Gemini is going to generate. So you can see right here that the script has been generated for me. So if you notice the ideas, no voice over and there is no dialog. So what gemni did write here, so generates a script that I'm just going to follow to generate my clips, my video clips using View three. No need for voice val generation, no need for adding dialog. Okay? So it I sit and you can see how different it is from the other two types of scripts, okay? So which type of script should you actually go for when you want to create your videos? The most popular one is the narrative Voiceover script. So you must have seen a lot of YouTube videos, like documentaries, you know, videos on past history. They are basically voiceovers there. And then people that are actually running YouTube automation channels, they quiz their videos, you know, using AI generated assets or stock footage from the Internet. And they simply just add their voiceover to the video. You know, their voice over narrating a story. This one is way, way, way more easier for you to do, than the dialog script. Why? Because all you need to do is to just go to VT generate your vis rouse, and then degenerate your voiceover, take it to Cap Code. You take the two of them to Capcod and then you edit them into a single video. Now for the dialog script, you need to take into consideration that V three doesn't really generate audio every single time. So there are sometimes you try to generate audio using View three, and it won't even work at all. It has happened to me before. And also, View three doesn't currently support the addition of first frame or last frame. We try to maintain character consistency using view three. You can't simply do that. And Vw three is the only model that you can actually use to generate the dialogue between the man and the boy. So basically, you should also go for the dialog scripts, okay? But now, another way you can actually make use of these dialogue scripts is by creating a multi speaker voiceover. Okay? So, this one is just similar to the narrative voice over a script. Or instead of just one person to be narrating the story, you're going to be having two people narrating the story. As they're narrating the story, they are going to be entacted with each other, you know, talking to each other, asking questions, asking themselves questions, too. You understand me? I'm going to show you Nanna lesson how exactly you can do this, okay? So once you generate your multi speaker voiceover, you can easily just generate the sens using your two so for recap, there are three main types of scripts that you can use to generate your theme, right? The first one is the narrative voice over scripts. The second one is the dialogue voice over scripts or the dialogue scripts. And the last one is the visual only script. So right now, I want you to go to Gemini and first of all, pick a storyline that you like, okay? You can even use the story line that I used in this lesson. You can also check the prompts pack and copy the prompts that are used for each of these scripts while you're trying to generate realistic AI movies. Then after generating these scripts, I want you to just take your time and compare the three of them. Okay? Just compare the three of them and see for yourself how different they are from each other. For your final projects, you are going to be making use of the narrative voice over scripts. So you had better just copy the narrative scripts that you generate using Gemini. Okay? So you copy it, and then the next lesson, I'm going to be showing you how you can use that particular script to generate a single speaker voice over and also how you can work with the dialog scripts in order to generate a multi speaker voice over, seeing the next lesson. 17. Creating a Single-Speaker Voiceover: This lesson, I'm going to be showing you how you can generate single speaker voiceovers for your videos. All right, so let's get started immediately. Now, the two we are going to be using to generate our voiceovers is not going to be 11 labs. Okay? We're going to be working with Google AI Studio. So this is the interface of Google AI Studio, as you already know. So if you come down here, you're going to see URL contexts to live audio to audio dialog, Native image generation. And native speech generation. So this is what you are going to click. This is what we're going to be working with. So just tap it all left, click hit, and this is the interface that you're going to be met with. Okay? So, as I said earlier, we're going to be generating single speaker voiceovers first. Okay? So in order to do that, just come to this part here. By default, the multi speaker audio is selected. So what you need to do is just let's click this single speaker audio and you'll see the interface change to this. All right. So right here, we have the place for the name of the title and description of the work or projects right now. And right here, we have the Style instructions. Okay? We have the style instructions, and we have the text. When 11 labs first came out, because it was 11 labs I used to use for generating my voiceovers, when it came out, it never had this style of instructions. I'm not even sure if it has right now, but this one is a plus for us right here in AI studio. Okay? So if you come here, you're going to see the model setens. And under this model setens you're going to see the temperature slider. I already explained the use of this temperature slider in the previous lesson. So if you don't understand what this means by now, you are wrong, you're on the wrong path, okay? So you can adjust the temperature here. You can take it closer to two or put it at two. Oh, you can reduce it to zero, bring it closer to zero. And I said in the previous lesson that the temperature slider is simply used to control the amount of craziness in the outsput, okay? So that's it. And then we have the voice right here. If you click this dropdown, you're going to see different voices that you can select. They are not a load as this in comparison with what 11 labs actually offers, this is not a lot, but you can work with this for sure. You can work with this, right? So all you need to do is to pick a voice that you like. If you want to know what the voice sounds like before you generate your voice over, all you need to do is to click the play button beside the voice names. Okay? So if I play this now, this is what you're going to hear. Ready to build something awesome today? You heard that. Now, you can also read the tags right here. You can see this one is bright and higher pitch. Okay? Here we have orbits, middle pitch, informative lower pitch, firm, excitable. So you can read the tags. This helps you to actually make a good selection. So let's see this knowledgeable. This knowledgeable should actually work for this kind of scripts or this kind of voiceover that I'm generates. But it's for the boy that got lost in the desert. So we actually want a solemn sounding voice. So I think this knowledgeable would work. And there's this other voice that I use often this Encelads or clados. I'm not sure of the pronunciation. So this one actually has a breath lower pitch. Ready to build something awesome today? So you can see, just for the sake of this lesson, I'm going to select this Sado Tager. So selects any voice that you feel would actually work best with your video. Okay? Now that we selected the voice, I will set the temperature, we just need to go back to Gemini, go back to the scripts that Gemini generated for us. So this is one of the narrative voiceover scripts that Gemini generated, and rights is a style. You can see it in brackets, copy this and come to AI studio then you paste it right here. Okay? Read read aloud in a gentle melancholic tone, right to Style instruction. That's the style instruction. And then I'll go ahead to copy this block of text, which is for the voiceover. Okay? So once I copy this and come back to AI Studio, all I need to do is to paste it in this text box. This space for text right here. Okay. And then I'll just hit this run button. It's as simple as that. Generating your voice over is very, very easy. If you want to stop it, you can easily just click this to stop degeneration, okay? So it has generated the voiceover for me already, and the voiceover states 3 seconds long. Let me please now so you hear what it sounds like. They say a person can get lost in their own thoughts. But the desert it was a word he whispered to the wind. Alright. So you can see how easy it is. Now that we're done generating the first part of our voiceover, all we need to do is click these three dots right here. Also click it, we're going to see this download. So just hit the button, the download button, and it's going to download the voiceover for you to download it to your storage. That's it for the single speaker voiceover. If you've been using 11 labs to generate your voice overs, it's okay. If you took my course on shorts video editing, using cap codes, then you will notice that in that course, the voiceover section was centered around 11 labs. It was 11 labs that I used to teach people how to generate a voiceover, okay? If you've not watched that course, if you've not taking the course, you can just go to my profile and check it out. It's called the Shorts Video Editing Mastercas Mastering cap code Editing, okay? It's there, and you're going to learn a lot from that course, right? So you can use 11 Labs. 11 Labs requires credits for you to be able to generate voiceovers. And once these credits are depleted, you need to buy more, okay? And there's a subscription plan. But Google AI Studio is completely free, and actually just recommend it to you, okay? It's perfect. In the next lesson, I'm going to be show your house generates more Thai speaker voice overs using Google AI Studio. See you in the next lesson. 18. Multi-Speaker Voiceover: Now we're going to be creating the multi speaker voiceover. Okay? So all we need to do is to click this right here and it's going to switch to this interface. Now, creating the multi speaker voiceover is actually easy, but you need to take note of the format. There's a particular format you need to adhere to before Google AI studio would recognize your script. Okay. So first of all, let's go back to this Gemini and search for the Turbo for the moti speaker uga scripts that we generated. Okay? I think this is it here. This is it. So no, this isn't it. I need to scroll down a bit and I'm going to see. So I couldn't find that format one that I generated, so I generated a new one, okay? And this is it. So you can see the format old man, Leo, old man, Leo, old man, Leo, Old Man, okay? So I'm just going to copy this right now and come back to AI Studio, okay? So here, I can easily just just come here and select this, remove it. Okay? And if I paste this, this is what it's going to look like. Now, look at this O. You can see right here. It says, no speakers detected. Please ensure your scripts speaker names are also set in the right side bar. Also, notice that as I pasted the script right here, it appeared under the style instructions. Okay? It appeared under the style instructions, which is not meant to be like this. Our voice over speakers are supposed to appear on a different section right here. Okay? So now I'm going to delete this right now. So see this box right here for the style instructions. I decide to just put in the style instructions right now. I see the sound Something just busy just for this lesson. They sound or they took in a hushed manner. Okay. Let me just leave this. So you can see that it's also typed out right here. So that means if I also type out the style here, it's going to be filled in right here. But for dialog, okay? So a dialog, all I need to do is to press this plus button. And if I press this plus button, you see what's going to add. It's going to add picker one right here, and the picker one is also going to be added right here, okay? So here, I can now decide to type high little boy. I can see the way it's also typing it out right here. So do you notice something interesting? If you've not noticed yet, I'm going to explain right now. So whenever you want to create a dialog, it must have speaker one and Colon. Okay? So this speaker one is going to match the speaker one that you see right here before AI Studio can recognize it as a dialog. Okay? So that means you would have to be imputing, manually imputing the text that the speaker one is going to say as speaker too. So let's say the speaker one is the old man, right? So typing the text that the old man is going to say, the we press this add dialog. They will type in words the boy is going to say. Then add dialog. You're going to type in words the man is going to reply with. You understand just like that, but this is so stressful. Let's assume we have a long script. Are we going to be doing this one by one? Ah, so this is what you're going to do. Now going to click any of these buttons by here. Okay? So you can click this and it's going to autofill this section right here. Okay? So what you're going to do is just copy this. And then we're seeing boga consumer preference. Format, the voice over script you generated earlier to match the templates above. Simple. So you can see the proms that I'm going to fied Gemini with right now. I'm tending it to format the voiceover script that is generated earlier to match the templates above, this template right here, okay? And I also told it to not include the scene details. Okay? I just need the voiceover only formatted to match templates that I'm giving you, o? So let me just say this now and see what Gemini is going to do for us. All right, so it has been generated already. And from what I can see right here, it's only added speaker one here. I think the issue is because I did not specify who speaker one or who speaker turning this prom. So I'm just going to pass these forms here in and speaker one should be the the old man Speaker two, the lost boy. Okay? All right. So let me send this to Gemini now, and let me see what it's going to do for me. Hopefully, this prompt works. Good, good, good, good, good. So you can see, what it has done. You can see now. So all I had to do was to specify that Speaker one should be the old man, and Speaker two should be the old woman. It's just because I wasn't specific that Gemini actually generated everything under just one speaker. Okay. So you can see the way I did it. All I have to do now is to copy this response, and I'm going to come to AI studio once again. I'm going to paste it, and then we'll remove this. Okay? I'm going to remove this and remove this, remove this. And for the style, I would just say so you can see now. Can you see the difference? This particular scripts that I just posted there actually worked. You can see the way it's auto filled the boxes for each speaker. The speaker, one speaker to speaker, one speaker to speaker, one speaker too. It's auto filled everything so that we don't actually need to start adding it manually here. You can see? So whenever you generate the scripts for a multi speaker voiceover, make sure you tell Gemini, I'm still going to include this prompt in the prompt pack. So it's very important to download it and check it out, okay? So, make sure that you use this prompt right here to format the scripts that Gemini is going to create for you. Okay, use it to format it to have Speaker one and Speaker two so that Google AI studio would be able to recognize it. Okay? So it's as simple as this. And also, I forgot to talk about this in the previous lesson. Click this drop down, you're going to see Gemini 2.5 pro preview and Gemini 2.5 flash preview. Obviously, the 2.5 f is better, okay? So you just select it. Leave it in 2.54. You don't really have any business right there, okay? So now that I've generated my voiceover appropriately, all I need to do is to hit this one buttin and it's going to generate a nice voiceover for me. Slowly, child. The desert makes a man thirsty, but a flood can be as dangerous as a drought. Drink slowly. Thought you were a ghost. Ghosts do not carry water. They only carry memories. My name is Elias. What is yours? Leo, I I was with my family. We were camping. I just wandered off to see the big rock. It looked so close. When I turned around. When you turned around, the desert had moved. It does that. It plays tricks on the eyes. I've been walking for so long. I shouted until my throat hurt. I was so scared. Fear is a heavy stone to carry in the sun, but you carried it. You did not stay still. You kept walking. That is bravery, Leo. I was about to give up. And that is when I found you. The desert takes you to the very edge just to see if you will look for a way back. Now, rest. We have our own walking to do. Elias, do you live out here? All alone? Alone is a feeling for those in a crowded room. Here I am not alone. I have the sky, the stones, the wind. They are old friends. They do not talk much, but they are very good listeners. Don't you miss people? Sometimes, but people are complicated. The desert is simple. It wants only one thing from you. Respect. You give it that, and it will show you its secrets. You were brave, but you did not respect its distance. It taught you a hard lesson today. Is that my town? It is a town. There will be good people there. That's how you generate multiple speaker voiceover using AI Studio. It's very, very easy to do, and I need it to try it out right now. Experiment. The last scripts that I use for the single speaker voiceover, make sure you change it to a multi speaker format and then you format it to work with Google AI Studio, the same way I did it so minutes ago. 19. Synchronizing Voiceovers with Visual Content: Now we have a voiceover, we don't actually have visuals to complement the voiceover. So now what we're going to be doing is to start implementing everything that we talked about in the previous lessons. Okay? So the first thing we're going to be doing is to go back to Gemini as usual. Now, this is the scripts that Gemini generated for our narrative voiceover. You remember. So this is it. Now to generate visuals for this is very, very easy. You can see that Gemini also described the visuals for us, right? So what we need to do now is to just copy this. Okay? So this is the first thing, right? So what we're going to do is to copy this. All right? What we're going to do is to copy this right now. Copy this part. You can say that the visuals doesn't actually stop here. It goes like see everything here can't just copy everything and places in Google flow, right? So we're going to need to use our intuition. You need to use your head. You need to use your brain to break this down yourself, okay? So you're going to be taking small chunks, just little parts of the visuals, okay? And then you're going to be pasting it in Google flow. So Google Flow generates it for you. Alright? So you can see that this visuals goes with this part of the hold on, let me see. This is the visuals on the cinematic view. Oh, sorry, this is not the script. Let me just go up, right? I just need to go up. Still the same thing, though, right? So this is it, right? This is it. So you can say that the visuals are here, and we also have the first part of the voiceover. Remember, we already generated this and we already downloaded it. Okay? So we have this already, and this is the sin that complements this particular part of the voiceover. Okay? So what we need to do right now is break this down ourselves. Let's see. The sin opens with a wide sweeping shot of an immense desir I think you can actually copy everything here because there's not a lot. You can copy everything here and just go to Google Flow and generate the visuals. If we get the results we don't like, we can easily just split it and generate it in two parts. Okay? So let's go back to Google flow. And here in Google flow, we're simply going to paste this prompt. Okay? We're going to paste this prompt and make sure it's in view too fast. We don't need to use View three at all because we are not going to be generating audio. They can use the view too fast. They can even use the view two quality. Okay, you can use the Vito quality if you have enough credits to boon, or you don't really mind. But here, I'm going to be using the View fast. So I'm just going to has this button to see what Google Flu is going to generate for me. So it has generated the videos for me. Let me just check them out now. I'll play this one first of all. Okay. Okay. Hold on. Well, let me just read the scripts once again. Let me read the scripts once again. So, it says the sun opens with a wide sweeping shot of an immense desert. The sun is blistering white. It's a blistering white orb in the pale blue sky and sand dunes. And we see a small figure stumbling through the sand, okay? He is alone, close up, shows cracked lips. He's Okay, so and the way this thing is supposed to open, it's supposed to open with just a shot of the desert with no charter in it, nothing in it. The boy is not supposed to be in the shots at first, okay? So I should have read this first of all. If I had read this, I don't have pasted the whole prompt, but at the same time, V two did not even do it for me. So this bo we're going to do first of all. We're going to open we're going to call it this desert thing, okay? So I'm just going to copy this going to copy this let's copy this, first of all, stretch the horizon. Let's copy this. I'm going to paste it in Google flow to generate this for me. Okay? So I'm going to reduce this to just one out spilt causes it's just an environment. I don't really need much options. Great. So you can see my thoughts process. I hope you understand. So this is not really hard. It's not really hard at all. Once you already have your script right here, all you need to do is to create videos that are going to follow this script. Do you understand? The only reason I picked only this part is because it's because the video that Google Flu generated for me did not begin with a wide sweeping shot of an immense desert. Okay, I began with a boy, and I don't want that. I mean, that doesn't follow the script, right? We're working with a script. So you understand I mean, that doesn't follow the script, right? We're okay with the script, you know. So once this one has been generated, the next thing I'm going to do is I don't even need to generate this part again. I can easily use this order, the first one that was generated if the desert matches. You can see that the desert doesn't actually match. These two things are different. This one looks much more hyperrealistic than this one. They look different, though, okay? So first of all, let me play this Okay, great. Great. This is wonderful. So what I'm going to do now is to add to sing, okay? Add to sing. What I'm doing is not supposed to be new to you at all. Going to do add to sing, and then I'm going towards. I'm going to use the jump two feature. Okay, I'm not going to use the extend. Going to use the jump two. Okay? So it's going to select the last frame of this, the last frame of this video clip, and then I'm going to just come back to Jamii then I'm going to come here you see a small figoGod stuff for me, small figu young browns up stumbling through the sand. Alright, so I can copy this and come to Google flow and just paste this. Okay? So what I can actually do right here is I'm going to remove this whole path. So I can cut out this whole path and just generate only this video. Then in the next scene, I'm going to generate this part. Okay? So where he pauses and then he pulls out a one empty water boat from a small backpack. I can also do it like that. Okay? But let me first of all, send everything to see if View two is going to be able to generate it precisely for me, right? So let me just send this now. Alright, so this scene has been generated. Now, let's play this. I really hope this is good. I really really hope this is good, okay? Okay, well. Well, it did not actually generate the water bottle parts. So that means I'm going to have to spit it, right? So this whole process takes time. If you've actually read posts from AI filmmakers, you'll see that this whole process takes hours. It takes hours just because you need to keep trying to keep trat Okay. And in this video, I'm not going to be boring you with, you know, trying things, trying to get the right footage, you know, changing prompts, editing prompts, and doing whatnot. No, I'm not going to be doing that. So I'm going to do everything. I'm going to generate all these things by myself after this video, after, you know, once I stop recording this video, I'm going to start generating everything, and then I'm going to impose them into cap codes. Right, so all I want to achieve in this lesson is to make sure that you understand how you can generate visuals that match your voice over. All right? So it's the same thing. Once you're done with this one? You simply just scroll down and then go to the second thing following what we have in the script. Then you start working on it, you know, generating visuals based on what you have right here in the script. Not going to be something that you're going to do quickly, but at the end of everything, it's going to be worth it. So just take your time, right? Take your time. If you've not visited or you've not watched the projects video, make sure you watch your project video. This is the best time to start working on your projects, okay? So you can pick any topic from the three options that I give you. Pick a topic, generate the script, generate the voice over, then start generating the okay? So that the next lesson when we are going to be importing all these assets into cap cuts to start the post production process, you are going to be with me. Okay? We're going to be doing it together. This is a practical watch and implement class. So I want you to be watching this video and doing exactly what I'm doing at the same time. I don't want you to watch the video, and then later next two weeks, you start implementing it. No, do it right now so that you grasp everything completely, right? So I'm going to see you in the next lesson. 20. Importing AI-Generated Assets into CapCut: In this lesson, I'm going to be showing you how to import your AI generated assets into cap code. So let's get started immediately. If you don't have Capcde on your PC yet, you can easily just go to your browser and you search Capcde download for Windows or for Mac, depending on the kind of device you are using. Okay? So the first results, you can just click on it, and you're going to be presented with this page right here. DC download for free. If you scroll down a bit. I think you should see the option to download for MAC right here. Let me click this download for free. Okay. So once you click this download for free, you're going to say, let me stop this. How do I stop this has song? So I think this page is just for Windows, right? This page is just for Windows, so I can easily just come back here and hold on. I was thinking the MAC download is going to be here, too. So I can just collate this and and yeah, once I hit this. You can also download it from your app store and we can see the page. So there are different websites for it. So depending on the kind of device you're using, just search it Capcos download for Mac or Capco download for Windows and you'll be able to download it and install it on your PC. I already have cap codes installed on my PC already. This is it, and you can see the way it looks like. This is not the updated version, so your interface might look a little bit different from mine, but don't worry, it's alright, okay? So right here, we're just going to hit this button. This big creates project button right here. So once we hit it, it's going to open up the cap codes editing interface for us. Alright? So this is what the cacode interface looks like, and I see. I can simply just rename this so boy in the By in desert. So for this course, I'm not going to be teaching you everything you need to know about capos because that's not the main purpose of creating this course, right? So I have another course that's centered around cap codes, video editing is capos. And if you want to learn how to edit videos using cap code, you can just visit my profile and check out the course right there. It's called Shorts Video Editing master class. Editing shots videos, creating shots, bir videos using Capcde. Just check it out there, and then you can go through the course right. So for now, I'm just going to show you everything that you need to know, the necessary things that you need to know, in order to create your AI videos, okay? So for this lesson, I said I was going to show you how to import your assets. So you can just click this see this parts here. These drag and drop videos, photos, and audio files here. It's right under this Import section, right? So, make sure you're in this import section. You're going to be here by default. Also your interface might not look like this if you're using Calco for the first time. So this is what I did to make to put my player right here. I just hit this menu button here, this menu option, and then you come to layout. So in this dayout you're going to see the different layouts available. The vertical layouts is what I'm currently using. If you click this one, it's going to bring the player here, and this is what your might look like, okay? But I prefer my own player to be right here. So that's why I changed it right here and I change it to the vertical layouts, okay? You can change your vertical layout or you can keep using the normal default layout if you like it. So I can just open my AI gen videos folder. So you can see the scenes that I generated right here. I can just come Lefslick and hold and just drag it, okay? It's no dragon. I can just drag it to the timeline or even to this place right here. So if I drag it to this way it's just going to it. It's going to add it to the imported assets. But if I drag it to the timeline here, it's going to put it directly in the timeline and also in the imported assets. So you can just drag it to the timeline if you're going to be using the asset right now. Easy. So you can see it has inputed already, and if I just drag this slider or the preview axis around, you can see the video moving right here. Okay? And this is the assets right here. This is the scene right here. If you want to input more videos or more assets, more audio files, you can easily just click this Imports here. It's going to open up your file manager so that you can select the resource that you need to import or the assets that you need to import. I'm just going to come to this sins folder and I'm going to import every other thing. I'm going to select this by holding going to select everything by holding control while I click. Okay? So if you hit this one, I hit this one. It's just going to select this one by one. But if you hold down your control or your command key, if you are using the Macbook, you see you'll be able to select different things at the same time, right? And just click this open. Now everything has been imported to cap codes. So that's how you import your assets into cap codes. And the reason why I'm using Capcod is because I want to merge the videos and sync them with the voiceovers, okay? So I'm basically using cap code to create the whole movie. I have all my sense generated with Virtu, and then I have my voiceover. Let me also import that right now. I see it has been imported. This is it. This is the voiceover and just drag it down to my timeline and drop it. As simple as that, very, very easy. So now let me play this. Let me play this. In the middle of the Sahara, a young boy was lost. Lost in a sea of sand. That's it. So that's how you import your assets into Cactes. In the next lesson, I'm going to show you some basic features that I need to master in order to edit your AI generated videos using cact. 21. Basic Editing With Capcut for AI Movies: The first thing you need to do before you start editing using CAP code is to set your short code. You can do this by hitting this short code button right here. You click it and it's going to open up the short cut panel for you. All right? So the basic shorts codes that you need to configure right here are the split the zoom zoom in and zoom out. This select I think, these are the default short codes, and you can just leave them like this. But this splits I think by default it's B letter B. But I always change it to S. I change my to S because it's just better. It's closer to my left hand. Left hand side of my keyboard and yeah, splits begins. So you can say this as S, okay? And then for the select modes A, right? The splits mode, you don't really need this select left word, select right word, okay? And then this zooming and out is very, very important. For this zooming and out, make sure you use the up arrow button and the down arrow button. Okay, so Zooming takes the up W button, the upwards arrow button, and Zoom mouse takes the downward arrow button. All right? If you go to this basic parts, the basic section, you're going to see this delete Make sure it has its backspace, right then undo Control Z, reset or redo Control Shift Z, okay? Imports media, Control I. Also, as you're configuring your shorts cd, you can just use the opportunity to familiarize yourself with the shorts cut right here. So just scroll through and see what you can do, right? So it makes editing so much faster. Easier. All right, so that's it, and I'll just hit this button to cancel it. Now, the first feature I'm going to be demonstrating to you is the split feature. So I'm going to be working with the audio. Now, this audio, this voiceover that I got from Google AI Studio, I can't just use it like this. I can use it like this technically, but I would love to eliminate some long pulses. So if you listen to the voiceover that you get from AI studio or you might notice that some pulses might be extra long. You can also notice it in the waveform right here. So this is called the waveform. This audio you can see those lines right here, called the wavefoam and looking at it actually gives you an idea of what the audio is going to sound like, where the pauses are, where the loud parts are it. So you can see here that we have a relatively longer pause, like in comparison to the pause that we have here and here. So if you want to reduce this, we can simply just come here, put our playhead right here. This is the playhead. This line you can see right here. And this yellow line that you can see, as I'm moving my cursor around, it's called the preview axis. And you can turn it on right here, turn it on. So if you turn it on, it allows you to preview the video without moving your playhead. If you turn it off, you won't be able to preview the video. So it's very useful to always have mine turned on. And also, you can just hover your preview axis, over the video and split. So I split this audio right now by using the S short code. If you don't use the short code, you can easily just come here. You can see the split. If it hit the button, you are going to split the audio for you. Okay? And the only reason why it splits this audio instead of this video that is at the top is because I selected the audio. You can see it's selected. You can see this white line around it. The audio is highlighted. If I hit this video instead and I hit S, it's going to split the video. Now, what if I try to split without selecting anything? So let me just hold my playhead here and hit the split hits. So you can see what just happened. It splits the video instead. But not just the video. It splits what is on the main track. So this is called the main track in cap code. This is the timeline editor. This is the main track, okay? This is the main track, the default track. So if you decide to add another video at the top of this track, it's This video is not on the main track. This video right here, the one that has this cover in front of it is called the main Track. Okay? So I'm just going to Cont Z and we are good to go. So now, let me split this audio just a little bit. No. Just split it just a little bit, and there we go. Now, if I want to delete this part of the audio, all I do is to hit backspace, and it's going to delete it. And I can easily just let's click this part of the audio, hold it and drag it so that it met this part. And you can see how we just shortened it, how we eliminated a little bit of puse from this audio. Now, if I want to trim this video, you can see this let me use this first of all. Let me reuse this I'll use this, okay? You can see that it's just a video of this desert, right? So if I feel like this video is too long and I just want a little part of it, maybe the first 5 seconds. I just want the first 5 seconds by here. Then what I can do is to trim this video. So if I bring my ca, so if I hot at the end of this video, you can see the way it changes. Now I can easily just drag it and trim it until I hit this playhead right so that's how you trim your video or your audio or any other scene that you can have on this timeline because it's not just for Vg audio, can also trim effects and all that sends, okay? So that's how you use the trim feature on cap code, right? So I'm just going to hit Cont Z to bring this back. Now, look at this part. We have some options right here. We have this main track magnets. What does this do? Let me demonstrate it to you. So if I split this video and I try to drag this right here, you see what happens. It automatically goes back. You know, this particular video attracts this one back to it. And that's why it's called the magnets feature. And remember, this is the main track. It only works on the main track. Now, if you turn this off, you'll be able to drag this to any part of your timeline, okay, without having it snap back to meet the Okay? So this feature is very useful and you need to know if you don't know how to use carpets and you try to split and move some items on the main track, you might just find that the clip keeps snapping back to this one. So if this happens to you, you know what to do. You just have to turn off this, and then you can drag. Okay? If you can drag, you can drag. Okay? So that's it, what turn this back on. Now we have this auto snapping. So this auto snapping feature is very, very useful. You have to have it turned Okay. Always have it turned on, 'cause if you don't have it turned on, you won't be able to accurately drag. Now, let me turn off this one, too. Oh, no need. Let me work with this audio. So let's as some trying to bring this one to this part. You see that it's not easy. I want this particular audio to join with the other one right here. It's not easy. I can't even find the ending part. If I leave it here now and I try to zoom this out, you can see that there's still space here. Now I will still have to drag this again. And then if I zoom this, try to drag again, that's a lot of work. Okay? That's a lot of work. That's why we have this auto snapping. So if you turn this on and you drag this when it hits this part, you see this blue line. You see that blue line right there, and then you can easily you can just release what you're holding, the assets. Okay? Now, if I zoom this, you'll see that it exactly it's touching this other one. Okay? So that if I also drag this now, if I bring this here, you can see that it's touching this, see the blue line. So that's what the auto snapping is for. It's a very, very useful feature. Always have it turned on. This linkage, don't you don't need to. Don't it. Okay, I'm not going to delve. I'm not going to explain every feature right here. I'm just going to explain only the ones that I need. This is also the zo panel, you can see, so it's easier for you to just use the arrow up and arrow down. And that's what I'm using right now. So if you want to arrange your videos, you want to arrange your assets right here, you can just use the drag and drop feature. If you want to bring this in front of this one, let me turn this off. All you need to do is to drag this. Okay? You left click it and hold it and you drag this. So you can drag it anywhere on your timeline. Okay? It's as simple as that. Also, if you want to generate captions for your video, for your air generated video, you can easily just come here this caption section. So once you hit this caption section right here, what you need to do is to click this generate. So if you want to generate a caption in English, it's already English is selected by default. If you also generates captions in a different language, you can select a supported language right here and generate your caption. Okay? So I'm just going to hit this generates right now and you can see that I started generating the captions for me. So you can see the captions have been generated already. And let me play this now. Let me this put this you can see the captions auto generated for us, and they're synced with the audio. You can also do some minor editing to your captions right here, I'm not going to go deep into this. As I said earlier, if you want to understand how to use Capcde completely, you know, for video editing, you can check out my course on Capcde editing on Editor is Capcde for this course, we're only going to be working with what you need for your AI generated videos, you know, just to do basic editing on it. Okay? So now that we've generated these captions, I want to work on this particular video sins that I have right here. Okay? So I'm just going to mute this with drag you see what I did right there. I just hover my course so over this line, this is the audio line. You will drag it down. You can see what happens here. The volume reduces. You can also do the same thing here by just increasing the volume here, maxing it out, reducing it to the minimum. Okay? So this is where you control the volume. Volume slider, you can also control it from here. Okay? So this is what I have. Let me come back to the input section and drag the second scene. This is the second scene. Hold on. Okay, this is scene two and three. So let me drag this now and put this right here. I generated this using View two using the jump two feature. Now, let me play this. See what just happened. Now, there are two scenes. There are two scenes in this particular clip, two scenes in this particular clip. But there's a problem right here. If I play this the way he stood up from the floor, I don't like that because literally he was standing right here. Okay? He was standing in this part of the video, and there's no place where there's no place where he actually fell down. You know, it's not shown in the video. But in the second clip, Video two somehow made him stand from the floor. So I want to cause that part out unless the video start from where he's working right here. So I'm just going to drag the preview till I find the sweet spot. Okay, so I think this is good. I'm going to just split it. I'm going to drag the previous houses to the part where he was still. Okay, I think this is good. Then I'm going to gilt it. Let me turn this on. And Great. I will still cut the spat out a little bit. Just cut the spat out and play this now. As simple as that, so you can see the use of the split and delete feature in cap code. Now, this caption right here, I can easily just drag it down here or I can even increase the size of this video. I'll just increase the size of the video a little bit so that it will hide this view watermark. You can see this view watermark. I don't really want people to know that I generated it with view, even though they might know, but I don't want this watermark to be here. So what I can do right now is just to selects everything. You can do this after you're done editing so that you won't stress yourself. Okay? Just select all the video clips. Once you select it, you put your playhead there, and then you can easily just drag it right here till the view is hidden. You can even drag it all the way till the whole video covers the canvas. So this is the canvas. This black powder, you can see the background. And normally, it did not actually cover it. You see have this bolder, this black bolder at the top and the bottom. We don't actually want this. Okay? We don't actually want this. You can still work with sets. You can even just drag your captions. You can drag your caption down here to be in this border. It actually works fine. It looks good like this. But if you want your video to fill up the screen, just drag it. Drag it till it fills it up. Simple. Okay? So this is one 30% 34% skill. You can also do the same thing here. You can see be dragging it. And it's 134 right. I can easily just edit it to 134 and boom. Now, I did not select this part when I was doing it. I can do the same thing by coming to this scale. Once I once I select this clip, I can come to this scale section and type in 134 and boom. You can see what has happened. So if I play this video now, you can see that it's full screen. It's a full screen video. Okay. So that's it. You don't really need much editing right here. Just arrange your clips, split them, and you add your audio, add your captions, make sure that you write space. Also, your captions, once you select this caption here and you come to this captions part, read it, read the text, sometimes cap codes might not be accurate in its transcription. So you can read it, and if you see any parts that is not correct, you can correct it right here. Just type. Simple as that. And also, if the line of caption is too long, example like this one now, you can easily just put your co at where you want to split it. Let's say I just want this path in a place that had no paths to show on the screen, and then this offer no direction to show on the screen again, not at the same time. Is the screen my case backspace to clay the space, then tap, enter. Okay? See what just happened. It brought this to another line. Now, if I play this, you can see as simple as that, if I want to bring it back up, I can just come his backspace, I need to bring this back up and then put the space where he space between paths and. You can do this for all the capsules. Check the ones that are too long, capsons that are so long, and start breaking them, breaking them into chunks. Chunks? Printing them into little pieces of captions. You understand? Right. So that's it. And this is how you edit your video using cap code. This is how you merge your videos using cap code. Now, that's three or four. I have I've added the fourth scene, the second and third scene already. And then I have the fourth scene right here. I can easily just drag it down. Good. So this is the fourth scene. Now View two as usual, added something I didn't want. I did not ask you to make this guy run. This is what I asked you to do. So first of all, made this guy run before doing the spots. Now, to remove this, I simply just splitted the video at the part. If you look at this, if you look at the video right here, you can easily see where you need to split. You see this is where he was running and this where's on the floor. Just right here, split the video and I go to delete this as simple as that. So you keep creating your scenes and you keep adding, adding splitting deleting, whatever you need to do just to make it look good. Now, if you play this video now, you play this part, you'll see the way it abruptly changes from 134, sorry? Yeah. It abruptly changes from this desert to this part. Oh, let me use this part here. Okay? So this guy right here, if I play this way just changes sharply to this other thing. If I don't want this, I want the transition from this sing to this thing to be spots. This is where transitions coming, and this is something very, very important. Okay? So if you hit this transitions part here, you can sit beside the captions, hit it, you're going to see different transitions available for use. Okay? If you scroll here, you see the basic blur mask slide clitch and overlay. So I love using the overlay transitions. I don't need something so flashy. I just need basic transitions. Now, look at this mixed transition. It's a pro transition to. If I drag this here and I put it right where this one ends, this clip ends and waves some begins right here, let's go, if I let's go of it, you see that it drops right here. As simple as that, that's how you add your transition. And now, if I play this video again, let me make this full screen. You can see the transition. You can see how smooth it is now. If I remove this and I play this, you'll see the difference. See the difference. Now, if I put this mixed transition back and I play video, you see? So it's blowed out of this one and blurred into it's fitted out of this scene and fitted into the scene. Okay? There are other transitions that you can use. There's this white slash transition, then and now they are all wonderful transition. You can see this white flash. See? So just play around here. Check out the transitions and use the ones that I need to use. You don't need to use flashy transitions. Just use normal basic transition. You don't really need to do so much here, okay? So we're not looking to overdo while editing your videos. Just do what is enough. No more than enough. Okay? So simplicity is key. And also add transitions to captions, but not this way. If I try to drag this mix and I try to put it on this caption, you see what will happen. Transitions can only be added to video tracks. But what if I actually want this caption to blend in, like this one to blend into this one? I don't want it to just change sharply. What do I I need to do so. If I select this caption right now, I need to just come to this animation section. Okay? This animation works the same way as transitions. So right here, I can just the in animation, the out animation. We have the in out loop and captions. So this is the outpat which is the last part, the ending part of the block of text or the block of caption, okay? This block of text right here, why the in part animation applies to this part here, the beginning of the text. Once again, the out animation only applies to the ending part of the text block. The in animation applies to the beginning parts. So this is the end part. I want to add a fade out animation. So I'm going to click this out and then have this fade out animation great. Then at the beginning of this caption, I want to add a fed in animation. Where is the fed in good. So this one is going to look exactly like this mix, because what this mix actually did is to fade out this part, the ending part of this sine and fit in the beginning part of this in. And that's what I just replicated here on the captions. So let me play this right now in full screen so that you see look at the trans. Look at the caption here. Let me drag it out. Let me drag it up here so that you can see it as well. So now let me play it. Look at the caption. You can see the way it's faded out and fitted into this. You can also adjust the length of the animation. If you don't want it to fid out too slowly, you can see that one was a little bit slow. You can adjust it and just reduce the time to 3 seconds here or less. Okay? If I do the same thing right here, click the caption and come here to the animation part and right here, sorry, it's in the out part, right? Okay, I can reduce this to 3 seconds, so it'll be a little bit faster. Let me clear this. You can see now. So that is for the transitions. This is just about everything that you need to know in order to edit your AI generated videos on cap code. Just split, delete, add transitions, add captions if you need to add captions. Okay, as simple as that. Now in the next lesson, I'm going to be showing you how to add music and ambient sounds to your video. 22. Understanding Sound Design: Now let's talk about sound design. What exactly is sound design in video editing? Sound design is the arts and crafts of creating an immersive audio experience that complements and enhances the visuals. So it's not just about the yarn sound effects. It's about sculpting the emotional heartbeat of a scene. In order to master sound design, you need to understand the different pillars that are involved. The first pillar is the fully effects. What are fully effects? Fly effects are everyday sounds we created in the studio, like footsteps, dog creeks, rustling clothes, just to match what's happening on screen. For example, if you have a fighting scene and you want to mimic the sound of breaking bones, you can just get pasta. You know what pasta is, right? Spaghetti, get like a pack, open it up, and then break the pasta in front of your mic. It's going to sound like breaking bones. Do you understand? So that's a fully effect. The whole idea is just trying to recreate, trying to mimic sound yourself. So instead of going online to search for the sound of breaking bones, you instead try to create that sound by yourself in a studio. So fully effects are very useful. If there's a particular sound effects that you can't find online, you know, on the free sites and you don't have a premium subscription, you can just use your creativity to recreate that sound using the things around you. Also, if you don't know what to use so we create a particular sound, you can easily go on the Internet or you ask had GPT or Gemini, for what you can actually use or what you can do so we create a particular sound. Another example is trying to recreate the sound of falling. You can just take rice and start pouring it on the tin foil. The sound that's going to generate would sound like a sound of rainfall or would sound like rainfall. Okay, so that's for fully effect. So the second pillar of sound design is ambience or ambient sound. I already mentioned this several times earlier in the course, so you should be a little bit familiar with it. Okay, so what exactly is ambience or ambient sounds? There are subtle background noises that sets the tone of your video. For example, the home of a city, the chirp of crickets, the sound of wind blowing, the crackling of a fireplace, stuff like that. So these sounds are not really pronounced. They are just the background in fact, if you are really engrossed in a particular movie you're watching, you might not even notice this ambient sound. So for this particular theme that I'm creating, an ambient sound that I could actually add, you know, in the middle of the sun. This opening scene where it's just showing the desert. I could add the sound of wind. If you have a particular scene where everywhere suddenly goes silent, you can amplify the sings by adding the chirp of cricket. And that's an ambient sound. Right? So now, the third pillar of sound design is the sound effects or Sex. Now, you must be very familiar with this one. What exactly are sound effects? Sound effects are more deliberate and exaggerated sounds like the sound of a sort clash, the Whoosh sound effect. You know, the popular Whoosh sound effects are a lot of people a lot of creators use. Yeah, that's the sound effect. Another example is the ringing of your telephone, the beeping of the fire alarm and stuff like that. So those are sound effects. Now, ambient sounds can be mistaken for sound effect, but there's a nice difference. Okay? As I explained earlier, ambient sounds are like background sounds. There are subtle sounds. While the sound effects are much more pronounced, they are exaggerated. You really hear them. I know that this is, you know, this is a sound effect. Ambient sounds just work in the background to help set the tone of the movie. Alright, so the fourth pillar of sound design is audio manipulation. What exactly is audio manipulation? Audio manipulation includes pitch shifting, reverb, distortion, and all that souls that seek recorded audio to create something new. For example, this is my voice over here. If we get to the parts where the old man is talking to the boy, and let's assume this is a horror movie. Okay? And the particular scene where the old man would be talking to the boy. So this boy were like, hey, let's assume the old man suddenly morphs or shape shifts into a monster. Now, if I want to mimic a monster's voice, while still preserving the old man's voice? I don't know if you understand that. Like, I want the old man's voice to sound like the voice of a monster. What can I do? This is where audio manipulation comes in. So you can tweak the sound, add voice changer effects, you know, to make the sound, you know, increase the bass, make it sound much more horrifying. So if you want a particular voice to sound like it's in a large hallway. Okay, let's assume your character is in a big building, an empty building. You want that voice, you know, to reflect the fact, that is in an empty building. Are different audio manipulation tools that exist, and audio manipulation is not a beginner topic, right? It's something that you really, really need to study. This is something that sound engineers. This is a specialized job for sound engineers, right? But you as an AI movie creator, can still work with audio manipulation, but basic audio manipulation. Okay, so cap coots actually has some basic audio manipulation features, which I'm going to show you later on in this particular lesson. Now, the last pillar of sound design is called layering and timing. This is the process of carefully stacking and syncing sounds to create, reading, build tension, or deliver punch lines. Layering and timing requires creativity. It's basically just stacking sounds on top of each other. You can see this download now, this caption, let's say, I want to put a sound effects. Let's assume this particular clip that I'm holding is a sound effect, and I want to edit on top of this where the guy starts talking. I can just put it here. So these are layers in this editor right here. These are layers. These are sound layers, audio layers, okay? This is just basically what layering is. And in the most basic form, what you need to know about layering and timing is just making sure that your audio really aligns with the video, and then you stack or you put the sound effects right where they are meant to be. All right. So those are the pillars of sound design. And now I'm going to be showing you where you can get your ambient sounds and your sound effect. So for the ambient sounds, the first place you can actually get them is right here on cap codes. Just come to this audio section right here, click it, and this is where you're going to be met with. So if we scroll down, you see the sound effect. You see these different sections right here. You have to import your music, sound effects, and copyright. What we're interested in is the sound effects. Just click it and scroll down. Once you scroll down, you're going to see this ambience section. So just click it and you can see the different ambient sounds that you can see shower water sounds, Boston street, office environment sounds, relaxing environmental sound, different ambient sounds that are available on cap codes for you to use in your project, right? So if you want to use a particular ambient sound in your project, you can easily just let me scroll off here. Let's say I want to use this sound of wind and ring, just hold it and drag it to your timeline. Okay, as simple as that. So this is one place that you can get your ambient sounds. Now, you can see that most of the ambient sounds even most of the ambient sounds you sound the **** right here on Cap pots are P. But I'm using the P. I have a CapcosF subscription, though, but most of them are P. Okay. So what if you don't have a CapcosPr subscription, but you need ambient sounds? Where can you actually get ambient sounds for free? This is where Pixar B comes in. So just visit pixab.com. Pixab.com uses the sites where you can get photos, illustrations, vectors, videos, music, sound effects, GIFs for free. Okay, so this is the parts where we're interested in the sound effects. If you click it, it's going to change here, and then we can search ambient sounds, okay? So what's to search ambient sound. You're going to see copyright free wind sounds, soundscapes. Look at them under water with ticking clock. So a lot of swing not just a lot, 20,811 yt free ambient sounds and sound effects available for use for free. Okay? So this Pixabay is a wonderful site. You can also search for specific ambient sounds. For example, the sound of desert wind sound of desert swings. If I search desert swing like that, you'll see. Let me play this now. See what that sounded like. We up to deserts wind. If you go down, strong howling wind, different ambient sounds available for use. Okay? So bookmark this site. You're going to be using these in loads as long as you are a video editor. Pixab.com, that's it. If you want to download ambient sound, for example, this Deserts wind, you can just easily hit this download button and it's going to download the sound effect for you straight to your storage. You can see it right here. Then you simply input it into cap coodes and use it in your video. The next pillar of sound design is sound effects. Where can we actually get the sound effects from? We can get them from cap codes or we can get them from Pixl B. So for cap codes, it's the same thing. Once you just click on the sound effects right here, you can see the different sound effects, trending, the new physical, phony, angry things and all that. So if you want the particular sound effects that is not really under any of these sections, yeah, any of these sections right here, you can easily just use the search bar. For example, we even need the Whoosh sound effects. I can easily just come here and search Whoosh as simple as that, you can see the simple woosh different woosh sound effects. And if you want to use it just drag and drop simple. Also, it's the same thing for Pixel B, but still under the sound effects section right here. If you click it, you can see the sound effects now. You just simply search for whatever sound effects you want, like the woosh sound effect, and it's going to bring out multiple woosh. You can see multiple Whoosh sound effects for you to download. If you don't find the one you like you can easily just scroll down and come to the next page. All right. So it's as simple as that. Now, to get music for your video is still the same process. For Pixel B, you simply click on this dropdown and you click music. You select music, right? So clay the woosh. If you want a simple background music for your video. Just search for it, and you get different music that you can use for your video. Now, you need to understand something. Despite the fact that this background music, you can see where Pixa B says the music is Woyalty free. All the music are Wyat free, right? But no, not all of them are WyatFree. If you're going to be uploading your video to YouTube, you need to be very wary of the kind of music that you use, right, so that you don't get a copyright strike now, whenever you see this symbol beside any music here, don't bother using it. This symbol is called the Content ID. As you can see here, the content ID is a digital finger printing system that can be used by content creator, so easily identify and manage copyrighted content on YouTube. Basically, what this means is that this sound is copyrighted. So avoid these sounds at all costs. So you can see this one now. This one doesn't have the Content ID symbol. You can use the sound on your video. Now, from experience, I can tell you that the fact that this music does not have the contents ID symbol beside it doesn't actually guarantee that the music is copyright free. I've actually used the music that did not have the Content ID symbol, and I still got a copyright strike from YouTube. Now, as long as you're using music from Pixar B, what you need to do is whenever you upload a video to YouTube, don't set the visibility to public. First of all, set this as an unlisted video and wait at least 3 hours. So within the 3 hours, if the video is going to be flagged for copyrighted music, then it to be flagged and you can easily change the music. But if you go ahead to publish the video as a public video, in fact, during the uploading process, you see that YouTube is actually going to check for copyright sound. And sometimes this has happened to me, by the way, YouTube might not find any copyrighted sound in your video at first. But then after some hours of publishing your video as public, thinking that you are actually good to go, get a copyright strike. So that's why I always advise that whenever you post the video on YouTube, leave it as an unlisted video so that you don't have to take down a public video just because of copyrighted sound and upload it again. Imagine if you've actually got a lot of views. So those views just go like that. No. Right, so that's how you get your music. You can also get them from cap codes by coming to this music section. And here, you see the different subsections right here, and you can easily just search for the same B cloud, music right here, or any other kind of music that you need. You can see the different music that is available for you to use. Your video. All right? Lastly, how do you know the kind of music and the kind of ambient sounds you use in your video? It's very easy. Come to Gemini. Now, look at the scripts that you generated using Gemini. Look at this script. You see this script that I generated right here. If I scroll down to the spot now for this one, you can see that it's generated the visuals, generated the narrator's voice over, and then it's also added a music and ambient sound suggestion. So for the music, it's actually suggesting that I use a sparse Huns Melody. Perhaps a single cello or a synth pad. And for the ambient sound, you can see the whisper of wind blowing across the dune. So for these parts, the deserts wind can actually work. You see, that's what I actually downloaded back then and I can easily just drag it to my timeline or I can just come back to this ambient sound. Sorry, it's under sound effects, right? And yeah, ambience just sage deserts wind. Good. You can see this. I can now drag this to my timeline and use it. Alright? So Gemini has already done the heavy work for us. All we need to do is just copy and paste, read and do. Okay? So for this music, you see, it's a musical instruments. I can still come to Piga Be and under music, I'll search Chu. You can see the different kinds of, you know, different music that I brought out for me, say, Cello melody so Cello. And it's the same thing. If you come to cap codes and if you come here to the sound effects, if you scroll down, you see musical instruments on that sound effect, not even on that music right now, on that sound effects. And yeah. So let me just search Celo here and see see different hello sounds. I still come under this music here and search Cello. Let me see if I can actually get something from those parts. So cap codes and Kiser B are actually enough for you to get your music and your ambient sounds, for your videos. If you own a large movie production house, I actually suggest you check out Epidemic Sound. So Epidemic Sound is like the best sound library available right now, and it's not free. The personal plan is about $7 per month, and the commercial plan is about $25 per month, but it actually gives you more rights to your music. It gives you all rights to apart any music you choose to use in just one license. So if you want to avoid all the hassle of getting copyright free music, you can try out epidemic sound. Okay? It's worth it. But if a small reto, Capote and pizza B are enough, alright? So that's it for this lesson. In the next lesson, I'm going to be showing you how to export and prepare your video for uploading. 23. Exporting AI videos: This lesson, I'm going to be showing you how to export your video and compress it before uploading. All right. So if you want to export your video from Cap Code, all you need to do is to click this Expose button right here. And if you use P features, make sure that you have a Cap Code Pro subscription. As you can see right here, I use two P benefits, but I can see exports because I have a CaposP subscription. So the first thing you need to do right here is to give this theme or this video a suitable name. So give your video a name that will enable you find the video easily. Now the next thing to do is to choose your exports folder. And this part is very necessary if you just installed car pots on your PC. So you need to choose an export folder right here. You simply click this folder icon here and it's going to open up your file manager for you so you can easily just select a fold that you want your edited videos to be stored in. So you can choose the fold that you created on the AIGN videos and choose this one and just select folder, right. We already have a folder selected here, so I'm not going to select anything. The next thing is to come down here. Make sure that this watermark is turned off. Make sure it's turned off. And then for this video here, you'll see that we have a box right here. So if you uncheck this box, it means you are not exporting any video. If you uncheck it and you scroll down and you check audio, this will allow you export only audio, only the audio file in P three format, as you can see right here. But if you want to exports the video only without audio. Most of the time, by default, the video and audio are going to be checked. So you need to scroll down and uncheck this audio. Make sure you uncheck the audio then you come to this video and check it. All right. So for the resolution, leave it at one ETP. For the bitrate, leave it at recommended codec, leave it at H 0.264, right? H 0.264 right here, and for the format, give it to the MP four. The frame rates, leave it in 30 FPS, right? Now, note this. A video you export and download from Google Flow comes as the 130 FPS video. So these are the normal parameters for the video. Now, if you decide to increase this FPS, this blames per second, so CC fs, you get this. You can see this message right here, this warning right here. So what this message is that you need to turn on this optical flow, right? So if I turn on this optical flow, see what happens. You see that the warning disappears. So what this optical flow does is that it matches your videos framewith to the frame with you choose while exporting. So what dismiss is that? If your video is a 30 fs video and you decide to choose 60 FPSA, you might not even see any difference unless you turn on this optical flow, right? So this optical flow is going to upscale or increase the framewth of your video to match the framew you choose right here. But the best thing to do is just leave the video framew at FPS and just leave this astra Cal flow turned off, right? And for this sync exported videos to space, don't sync your video to space. Why do I say so? The only reason why you would love to sync your video to space is so that you can work on your video from another device. Let's say you change location and you leave your PC at Ooma, you just want to work on your video from your phone. Yeah, you can access your video from the Spaceball. It's not even worth it. Because Capcos changed their terms and conditions, and now CaCOs has rights to any video you upload to their space. So they can decide to use it for anything TW. So they can decide to use it for anything, and you can't sue them. You can't sue them because it's in these terms and conditions that you agree to before you use the software. So just to be on the safer side, don't sink your videos to space, right? Now, if you want to export only the audio of your video, let's say you want to export it and compress it or you want to work on it with another software. This is where you need to check. So if you check this, you leave it to Empty format and it's going to upload. So if you check this, you leave it to p format, and it's going to export the audio for you so right, so the rest of the set is right here, you don't even need to check them at all. These export GIF captions, you don't need to check them. The only thing you're really concerned about here is the video, right? So once you're done, what you need to do is to click this Export button and to start exporting your video as simple as that. Now, my video has been exported already. The next thing to do is to compress my video. And what do I mean by compressing? Compressing is simply reducing the size of the video without altering the quality. Why do we need to do this? For example, if you edit a video or if you create a video that is about an hour long, the size of that video is going to be large, and it's going to take time to upload to the different platforms you want to upload it to, it's also going to waste data. So what you have to do before you start uploading your video is to compress it. And the tool that I used for compressing is called hand break. So this is handbek right here. And if you don't have handbek on your PC, all you need to do is to go to your chrome or your edge browser and search handbrake download. For Windows, if you are using the Windows PC or for MacBook, if you are using the MacBook PC. Okay. So this is handbrake right hair. You open it, this is the interface you're going to be met with. So what you need to do right here is just to drag and drop. You can drag and drop the video that you exported from calcs and, you know, drop it right here, or you simply click this file, these open video files, and then you sedex the video. And that's what I'm going to be doing right now. So before I select this video, I would love you to look at the size of the file. I said that it 7.1 MBB fore file, 37.1 B. So this is before compressing the video. After compressing the video, we're going to check the size to see if it's actually reduced the size. Alright, so I'm going to select this and open it up in handbek now, don't be overwhelmed. Handbrake is quite easy to use for compressing videos. First of all, we have different presets. This first one P 30 FPS preset can be used for this. Okay, so I can just click Starts code and it's going to compress this video for you. But what if the video I want to compress is not a system nine video? What if it's a nine system video? In this case, this one P thtFPs will not work. Okay? I need to create a new preset. And how do I do that? I simply click this save New Preset. Okay? And here, I'll just give it a name like shorts like, short, new. Okay. And right here in this resolution on limit, I'm going to click this drop down and select Custom. All right? Now, you can see that this is 1920 by 180. What this simply means is 69. So if you want to compress a 916 video, what you need to do is to swap this value so you need to put this 180 here. I put this 1920 right here. So I'll simply do that by typing 1920 and typing 180 right here. And that's all you need to do, simply click Add, and it's going to add the preset, right? So now, this preset is what I'm going to be using to compress 916 videos, right? But if I want to compress a video that's a 69, what I need to do is not to swap the values. So when creating the presets, all you need to do is also click the Add button. You're not going to be swapping any value right. And even here, if you click this presets right here, you will see different presets that you can choose from. So you can see this first. The first one CP. Look at this first one CP, 30 P is right here. If your video is not one T P 30, if it's 720 P, you can choose this. You can see different presets right here. So when you create a preset, you see it under this custom preset. I see I already have research for short videos and for long videos, okay? And right here, you can select this wave optimized. You can check this wave optimized button right here before you start encoding. All right. And then the next thing to do before you click this starts Encode is to choose an output folder. So you can see that I created a folder specifically for compressed videos, and I named this compressed handbrake. So if you hit this browse button right here, you'll be able to choose an exports location or an Exports folder. Now I'm just going to change the preset back to a preset that is suitable for long videos, so I can just choose this long new preset right here. Now, look at the size this is this storage and display, always make sure that the storage size and the display size are the same thing. So if the size of the video you import into handbrake is 180 times 1606 display, make sure that the storage size, which is the size of the outsource file, you'll be get also the same thing. All right. So let's go. I'm going to hit the start in good bottom right here, and it's going to start compressing my video for me. 24. Congratulations and Project Reminder: Congrats on completing this course. You are now an AI filmmaker. Congratulations. I am sincerely happy that you completed this course, and I really, really hope you enjoy this course and you learned new stuff right here. So just as a reminder, this is your class project, not forget to submit your class projects. There are three topics right here. Just choose one and create your video. Alright? Lastly, please leave a good review if you actually learn something new from this course. To the review section. Make sure you leave a good review and check out my profile for all the classes and courses that you can take. I wish you all the best.