Gemini 3 + Nano Banana 2 + VEO 3.1: Master AI Video Creation

1. INTRODUCTION: Imagine creating stunning AI videos without spending hours on editing and learning complex software. Now you can with nano banana and video. In this course, you will discover how to turn your ideas into AI cinematic videos step by step, from generating scenes with nano banana and refining them with view. I'll guide you through every part of the process. No more guesswork and no more stress. Pure creativity powered by AI. Now, whether you want to grow on social media, you want to build your brand, or you want to tell a better story. This course gives you the confidence and skill to do that. So what are you waiting for my friend? Join now and bring your imagination to life with the power of AI. 2. How to Generate AI Images: Hi, everyone, and welcome to this ultimate AI content creation with Vo. Now, first of all, I'm going to show you how we can generate images via the Google Image generation model. To use that model, you have to go to aistudioggle.com. You have to go to chat. Now, if you want to switch your theme from light to dark, you can go to home you can go to settings. You can hover your mouse to theme, and you can select dark, and boom, we have dark theme. Now, if you go to chat, we have several options right here. We have Gemini images, audio and video. For Image model, we can go to images. We can select any model. So I'm going to select Image and four. And as you can see, we are now using Image and four model. You can also switch your model from here. You can click here. And as you can see, we have Image three, Image and four, and we have the latest one Gemini 2.5, which is also known as Nano Banana. Now, if you don't want to use AI Studio, you can also use labs dot Google. This is the site labs dot google slashxTolla Image FX. As you can see, guys, the image FX is now using Image and four. In previous they used Image and three. Now they switch the model 3-4. You can generate unlimited images for completely free. There is no limit. You can use this site or you can also use the AI Studio. Now the process is simple. You can select your model. So I'm going to select this and I'm going to type a simple prompt. So in order to generate image, you have to describe your image. So for example, if you want to generate a glass of water on top of a table, then you have to write that specific image description. For example, I'm going to type a glass of water placed on a table. Now there could be any table, so you have to be specific. If you want to place your glass of water onto a wooden table, you can mention that. If you want to place that on metal, you can also mention that. So you have to mention every tiny detail that needed to generate our image. So I'm going to type wooden table. And if you press Enter, nothing will happen because you have to click on this Run button, or you can press Control plus Enter. And boom. The AS studio is now generating this image. Let's copy this prompt. Go to labs dot Google, page that same prompt, and now I'm going to click on this Create button. Now if you generate your first image, you will get this message, so you can also enable your Google Drive. So now we have this image. Let's view this in large view, and as you can see, we have this high quality image. If you are wondering why we are getting this kind of aspect ratio, the reason behind is this because we have selected this aspect ratio. So we have to select different aspect ratio. So I'm going to select this one which is landscape, and now I'm going to click on this button again. Let's go to Image FX, and as you can see, guys, we have the same exact images. You can also switch your aspect ratio from here. You can change it to square portrait, landscape, mobile portrait, mobile landscape. Now we also have this log which is called seed. Now, why this is important, as you can see guys, every image is different from one another. As you can see, guys, we have four images, four results, and every single result is different than each other. So if you click on this log, then it means that it will generate specific results according to this sed. Let me show you what I mean. So I'm going to click on this create again. Let's go to AI studio, and this time, as you can see, guys, we have this landscape so by using the AI studio, this is how you can generate any image that you want. Now let's go to Image effet, and as you can see, guys, if you notice, now we got similar result. Now, this seed is important because you will get similar result if you press this log. Now if you unlock your seed, then it will give you more variety outputs. So I'm going to click on this Create again. And now as you can see, guys, we don't have similarity in this result. Every single glass is different than each other. If we go back, you can also go back to your previous result. As you can see, guys, every glass is same in every image. We have slightly different camera angle, but as you can see, guys, every glass in this image is exactly the same. If we go back to previous result, as you can see, guys, without the sad, we also have the same result. But if you unlock your seat, you will get different result. If you lock your seat, then you will get similar result. This is the difference between the set log and log. Now if you want to generate images, you can also use a different site that also run by Google called WISC. So you can click here, and as you can see, we have WIX right here. I'm going to click here to access to WISC Let's Inter tool, and as you can see, we are using the WIS. You can also use the Wix for completely free. You can also generate unlimited images. Now like the image effects, we can click here and we can lock the seed. If you want to change your expect ratio, you can select your expect ratio from these three options. If you want to get some inspiration, you can click here. And as you can see, guys, we have some inspirational prompts right here. These are some examples that you can try on. You can also add your image. You can add your subject, add your scene, and you can add your style. So these are some different ways to generate unlimited free images by Google. 3. Free Ways to Access to Nano Banana: Okay. Okay. Hi, everyone, and welcome back. Now I'm going to show you how we can get access to nano Banana. We have multiple ways to get access to nanobanana. First of all, go to gemini.google.com. Once you sign in, we have to click on the tools, and once you click on the Tools, as you can see, we have this banana icon, and we have create images. Once you select this one, you can describe your image. So for example, I'm going to describe retrostyle car parked in a abandoned city. We have to click on the Summit button, and as you can see, guys, it is now creating our image. If you misspell something, then it will also correct your mistake. And now, boom, we have this image. We can simply go to the aistudio dot google. Once you get here, we can click on this try nano Banana, or we can go to chat. We can go to images and we can select Nanobnana. Once you select this nano banana, you don't have to do anything. We can just simply click on this nano banana and we can describe anything. So I'm going to copy my same prompt that I use right here, or we can click on this button. It will copy your prompt. Now we can go to Gemini and now I'm going to paste my prompt right here. I'm going to select this nano banana and now boom. Now check this out. Now we have another way to get access to nano Banana. Simply go to labs Google slash AxlaTollash WIX, G to Intertool and we can page that same prompt and we can click on this button. Once you click on this button, it will create your image. Now, once you have your image, you can click on this refine. And as you can see, guys, if we click here, it is using nanobanana. If we hol a mouse, as you can see, guys, this model is using nanobanana. Now we can describe any changes that we want. So I'm going to describe so I want to turn off the headlights of this car, and I just type make light turn off, and boom, check out this result. So this is before and this is after. And the best part is that you don't have any watermark. If you click on this download button, if I open my image, as you can see, guys, the image is in high resolution and we don't have any watermark. Now we also have another way to get access to nanobanana. Simply go to labsdGoglelashFx tool and slash FLO. Create a new project. If you have one, you can open your previous project. Click here and go to frame to video, click here and click on this generate button. Now I'm going to paste my same prompt, and now I'm going to click on this button. But before you click on this button, you can also click here and you can select your outputs. So currently, I am using four outputs, so I'm going to click here and let's click on this button. And as you can see, guys, we have these images. Now, if you want to use nano Banana, you can simply select any image, and as you can see, this model is using nano banana. So we can edit the image, or if you don't satisfy with your result, you can click on this Create. You can paste your prompt and you can generate more images. Now, this time it is generating one image because we have selected one output. Now, this time I got this image, so I'm going to select this image. So I have type, add rain to this scene, and now check out this result. So these are four method where you can access to nanobnana four completely free. 4. How to Access Nano Banana 2: Yeah. Welcome everyone to Nano Banana Two. In order to access to Nano Banana two, you need to go to flow. So search for flow, and you have to click onto this website. Now, once you open this website, you need to click here Create with Flow. Now, this is where we can access to Nano Banana two for completely free. Let's close this tab, and as you can see, I have several projects already. If you are opening the flow for the first time, you clearly don't have any projects, but you can create one by clicking onto this button. New project. Once you click on it, you have your first project. Now, right here, you can rename your project. You can delete it as well. And if you click here, as you can see, we can upload our assets. We can upload our media images. If you click here, you can access to Nano Banana two. You can access to the previous version of Nano Banana, which is Pro, and you can even access to Imagen 4. You can select your aspect ratio right here and you can also select the number of generation. If you click here, as you can see, you can also access to the VEO 3.1. You will get 50 credit per day, so you will get daily 50 credit for completely free. So this is the basic interface of Nano Banana two. 5. Basic Use Cases | Nano Banana 2: Welcome back, everyone. So now let me show you how we can remove background or even change it. First, you need to upload your image right here. Now, once you have your image like this, you can type prompt like remove the background. So I already done that, so I'm going to click here to reuse the prompt, which is remove background. Now, using the nanobna two, I got this result. You can also click here and you can see which model is used to generate your image. And as you can see, this is created on this date using nanopanana two. Now I also use NanobnaPro if I click here, as you can see, this is a nanobranaP and I got this. So in the first case, I got the mask which we can use to remove the background of any image. In the second, we have this transparent background, but it is just change the background to this transparent background. The image is not even intransparent. Let me show you. So you can also click A and you can download your image. Once you have downloaded your image, you can click here to open your. And as you can see, this is not a transparent image. The background is just a transparent background. Now, the best way to remove the background is to change not even remove it, change the background. So I have this example where I use the same image. If I go to the history, as you can see, I have used this image, and if I click here to reuse my prompt, you can even click here to reuse your prompt. And as you can see, I have typed person standing on top of the building with sun behind him. And as you can see, we have this beautiful image. Now we have the second example where I type, if I click here to show history, if I click here, person standing in front of a giant concrete wall. And as you can see, this is a pretty good looking so the protein is don't remove your background, manipulate or change your background. You can also remove particular things in your image. So for example, in this image, I want to remove this apple. This is the result that we got. As you can see, I have to remove apple from the scene, and if I open the image, it removed apple from this scene. So this means you can remove anything, any object. It doesn't matter from your scene completely. Can also remove noise or grain from your image. As you can see in this image, I have so many grains in my image. So I use nano banana two. As you can see, I've used nano banana two, and I have to remove noise and grain from this image, and it removed the noise and gain, but it also changed my jacket and the background as well. So I use nanobnanaP, and as you can see, nanobnanaP result is more convincing because we have the same background, but it also changed my jacket as well. This is the second result. Can also change the color of any object of your scene. So for example, in this scene, we have this apple. So I just type change color of the apple to yellow, and as you can see, it beautifully turn this apple into yellow color. You can also colorize your black and white pictures. So I have this picture, and I just colorize this image, and as you can see, it colorize this image. Now, the best part is I lower my saturation to 0% of the image, and I uploaded that image, which is this one, and then I colorize the. Let me show you the original colorized result of this image. So this is my original image result with the color. And without the color, this is the one that we used. So I think it really did a pretty good job right here. The colorization was pretty much incredible on the landscape. Let me show you how much accurate it is for your subject. So as you can see, we have this image of myself, and with the nano Banana two, I type colorize this image. And as you can see, we have this colorized result. Now, this is not the original colors of my jacket, but I think it really did a pretty good job. Now, this is my original image with the color, and I lowered my saturation to 0%, then I use this. So as you can see, this is the original colors. But the NanomT did a pretty good job on guessing which color of the object or the shirt jacket is. Not only you can upload your image then manipulate it, you can also generate your image from scratch. You can see, I have type luxury couch, and this is what we got. Then we have this example. I type abandoned rusty car, and this is what we got with nanopnana two. Then we have this alien like dog, and finally, we have this rusty bike. Now, you can even do basic editing like if you want to flip your image horizontally or vertically, you can type flip image vertically or flip image horizontally. Now sometimes you will get this kind of result, but you can re upload your image, type your prompt, and then you will get some better result. But after this prompt in Nanobnana two, I still got changes that I don't want to make in my image, like we have the scar and we have this light leaks effect from here. Now, you can also rotate your image by typing rotate image 90 degree or you can type any rotation degree angle. And as you can see, it rotate my image 19 degree. But as you can see, guys, you can tell we have some other changes as well. So next what I did, I used my original image, and I type rotate image 90 degree, don't change anything else. As you can see, it rotate my image 90 degree perfectly fine, but as you can see, we have some weirdness in the image. The reason behind why we have this because in the original image, we have the landscape image, and in this image, we have this portrait result. So it adds this transform effect. Let me explain what I make. So we have this editing software. Now, if you want to zoom in Zoom out, we have this effect. If you want to rotate your image, let's say 90 degrees, so I want to type 90, and as you can see, we have the 90 degree rotation. We can do this in minus, so we have the other way Now, as you can see, it also crop some part of my image. If I click here, as you can see, this part of my image is outside the main frame. In order to fix this, we can either crop this image or we can unlink the Zoom and we can do this instead. That's exactly what Nanobanana did. It just add this transform effect, unlink the zoom and add this transform. Check this. This is what it actually did to my image. Another use case is you can replace your person with another person. As you can see, guys, I replace myself with this three D boy. So first, I generated this image using this prompt pixel style three D boy in a jungle. If I go to history, then I use my image and I use this boy image and I say replace person with jungle boy and have a look. We have this perfect replacement. Another use case of nano Banana two is you can place any text in any style. So let me show you this example. As you can see, I have typed right in Cloud style text in quotation under we have welcome and I type behind Sky of this person. And as you can see, we have this welcome behind the have another example where I type write text inside concrete wall. In this under quotation, I have type welcome, and as you can see it type this text. Now you can describe your text font. You can describe your text property like your font size, you want to make it bold or italic, et cetera, and you can even take this to another level. This is just the basic practical use cases of nanobnana 6. Different Camera Angles | Nano Banana 2: Yeah. Oh. Welcome back, everyone. Now let me show you how we can generate alternative camera angles of our image. Now, as you can see, we have this image. If I go to history, if I select my image, I use this prompt, show low angle shot of this image, and this is what we got. Now, if you want to learn the camera angles and different kind of camera angle shot, now simply go to Google and search for camera angles, and as you can see, guys, we have these various camera angles that you can use in your prom. Now, this is my original image. And I type show white shot of this image. This is what we got. This is the white then I type show extreme close up of the person face shot of this image and have a look. Then I type, show bud eye overhead view of this image. This is the result. So as you can see, you can generate any alternative angle of your image. Now we have this Dutch angle, which means a little bit tilt effect. So I type show Dutch angle slash tilt shot of this image, and this is what we got. Then I type show over the shoulder shot of this image, and this is what we got. 7. Color Grading | Nano Banana 2: Yeah. Welcome back, everyone. So now let me show you how we can calibrate our images. In Nano Banana Pro, I have used this image and I have typed apply teal and orange color grade on this image, and this is the result that we got. Then I type, apply matrix color grade on this image, and this is what we got. Now, when you manipulate your image, you have to be careful because after this result, if you type any changes, it will apply changes onto this result image, not your original image. That is why we have this matrix result in different light. Like we have this light of this image in this result. Now, these are the nano BranaP result. Let me show you Nano Banana two. Now if I click here, as you can see, this is nanonna two, and I have typed the same prompt, apply teal and orange color grade on this image, this is what we got. I feel like the Nano Banana is a little bit better than the pro. Then I have used apply matrix color grade on this image, then I typed apply modern color grade on this image, check this out. Then I have type, apply blockbuster action color grade on this image. After this, I have type, apply dark and moody color grade on this. Now, it is applying these changes onto this first result image, not my original image. Let me show you this result and after that, I'm going to compare these with the original image manipulation. We have this pastel and soft on this image, and we have the final neon cyber color grade on this image. Ha. If you look at the dark and moody Colgraate, this effect is applied onto this image. If you have used your original image, so this is what I mean. If I click here, as you can see, I have used this original image, then I have typed this prompt. Apply dark and moody cologrd on this image. Check this out. This is the result of the dark and moody original image, and this is the result of dark and moody applied onto the first result and now have a look. Now we have this final spilt complimentary color grade on this image. If you're wondering where I got these color grade prompts, you can simply go to Google. You can search for color grade styles, and as you can see, we have these several styles, and this is where I got my style. So this is how you can color grade your images using Nano Banana too. 8. Copy Any Style | Nano Banana 2: Welcome back, everyone. Now, let me show you how we can copy any style of any images using Nano Banana two. So go to pintre.com and you can get your reference style from here. After you get your style, select your image, upload your image here. Now, once you have your image, you can upload your image plus your reference. If you have your image first and reference second, then you can type this prompt copy style of the second image and apply it to the first image. You have the reference first, then your image in the second place, then you can swap the word. Like you can type copy style of the second or the first image and apply it to the second image. Now we have this example. We have this reference. We have my original image, and I type copy style of the second image and apply it to the first. I also type this detail. I want to add some glasses on my face, so I type add ama, also make person wear black. If you want to make several changes, you can add a comma and you can describe your changes one by one. We have this example where we have this reference, copy style of the second image and apply it to the first. And as you can see, we have this style applied onto this image. So you can basically copy any style of any reference image and apply it to your own image. 9. Change Poses of Character | Nano Banana 2: Welcome back, everyone. The most creative use case of banana is you can change the pauses of your character. As you can see, I have this character. We have this pause, and I use this image and I type boy is doing pause as you can see the boy is doing the pause, but the pause is not correct. In the reference, we have a different pose and in this image, we have a different poses because we have a different camera angles. Now I use this on myself and I got this result. Then I use the same reference and I regenerate my image. Tas time I got this result. I use this reference image and I generate this image which is not a good looking image. Now I have used this vector style image as a reference for better understanding for the AI, and this time I got this result. This is not the post that I've posted. So I did I swipe my model to Nano Banana Pro and Nano Banana completely changed my character into different style, but it nail the pose of my character. So what I did, I run this prompt multiple times using different model like Nano Banana, Nano Banana two, and at the end, I have my pose. So if you are not getting the result that you want, try to regenerate your images because in this way, you will get your desired result. As you can see, the boy is doing the correct pose after I run the same command for five to six time. 10. Logo and Graphics | Nano Banana 2: Yeah. H. Welcome back, everyone. You can also edit graphics. As you can see, we have this billboard, and using the gemini do google.com, I have tie change the text to eat the billboard. And as you can see, replace the text. Then I tie M Billboard logo upside down, also change tagline to I am hating it. A all right meat chicken to vegetarian. So this is all white chicken. And as you can see, it make those changes that I asked him to make, but it did not change the logo upside down. So I say change McDonald logo to upside down. We have the logo. I type changed the burger, so I meant to save the burger on billboard to a bunch of salad and have a look. Then Is change the color palette of this billboard to blue and orange, blue and orange. Then Is now change to green and blue. We have green and blue. So this way, you can easily manipulate any graphic. Not only the graphics, you can also manipulate logos. So we have this logo and I say, make this metallic logo. Then I tie, make it more metallic and rusty, adjust the texture and match style. Have a I say, now apply volcanic style. Then I use this logo and I say now apply the same style to this. Now, sometimes it fails to generate your result, but you have to retry. After some reputation, I got this result, which is better. Then I use this logo and I say, make it fire logo. Then I say remove white stroke from this logo. Have a look. We can also apply any other style to our logo. So we have the simple FA logo, and I say make FA logo look like the right Image style. So this is the style. You can also go to printer.com to get your reference style and check this out. We have this style. Now we have this one. We have this one, O L, and finally, we have this one. So this way, you can easily manipulate graphics and logo using nanobnana. 11. How to Do VFX | Nano Banana 2: Welcome back, everyone. So you can also do the VFX using Nano Banana too. So I have this image of myself and I type person wear black sunglasses standing in front of a black wall. Then I use this and I generate this image by typing person firing machine gun at robots in a wall. And have a look. Then I use the same image and I typed this prompt, and this is the result that we got. If you're thinking about where I got this prompt, simply go to astudo.google.com. This is completely free. You can sign in through your Gmail account. This is the Gemini 3.1. I've typed this prompt. Give me 50 VFX idea for a person. And as you can see, we have these credible ideas. So I use this prompt for Nano Banana two. As you can see, we have several VFX shot. These are my prompt. These are the AI Studio prompt that I got from Gemini 3.1. Have a look. We have this example. We have this effect. We have this one, and final, we have this. So you can generate credible VFX shot using Nano Banana two. 12. Use Drawing | Nano Banana 2: O. Welcome back, everyone. Now, instead of writing prompt from scratch, you can visualize your final result. So I have this image. Let's click here and I'm going to draw onto this. I'm going to draw my glasses. We have this. Hello. Let's type. La glasses. Now, once you draw onto your image, you can simply type make changes to this image according to the instruction. I'm going to select model two NanopenNI. Let's generate this image. While this image is generated, we can go back. We can look at some other result. So I have used Window picture editing software to draw onto the as you can see, if we look at this image, this is my drawing. As you can see, the drawing is very rough. I tie make changes according to the instruction, and this is the result that we got. This is not what I was expecting. Now this is my second image. If I click here, as you can see, guys, we got this image. Now in this reference image, we have keyboard floating in air. I have drawn this keyboard. We have this haze. We have mouse floating in air, this out, and now have a look. This is the final result with nano Banana two. We have this example. So I have a little bit clean drawing. In this image, we have haze, we have mouse, and we have keyboard and check this out. Now in this example, I have type make changes according to the instruction. If we look at the instruction, check this out. We have this dust, and we have this spaceship drawing and check this out. We have the second variation. By the way, if you use your Window software to draw on your image, you can simply upload your image and you can add that to your. Then you can make changes to your image. This way, you can generate up to four images at the same time. Now we have this result. As you can see, if you look at the image drawing, we have this car and we have this dust and check this out. This is the second variation of the same prom. Now next we have this image where we have the line drawn. I arrow this up and I say, and now if you look at this result, I have type the symbol prom, make changes to this image according to the instruction. This is my second variation. And in this image, after I got this result, I have to field ends in the background and on the table as well, person wear face mask and check in this example, I have typed just make changes to this according to the instruction, also person a Face Mass. I really like this line holding onto me. Now, these two images are ready. As you can see, we have this creepy smile and these weird LEDs. So I have using the flow platform to draw onto this image. I got this trash result. In the second result, which I use the already drawn image, using my Window software, you can use any Window picture software to edit your image and you can draw onto result is in high definition and in the lens. If you look at the result, as you can see, guys, everything look crisp, clear, and fantastic. Whereas in this picture, it looks like we are using the early stages of AI image generator. So this is how we can use Nando banana to generate images using drawing. 13. How to Remastered | Nano Banana 2: Y. Welcome back, everyone. Now, another fun use case of Nano Banana is you can remastered your old stuff. So we have this game screenshot, and I have typed make modern remaster of this and have a look. We have the modern remaster of this then I use this metal slug screenshot and I make modern mustard of this. If you are not getting the result that you wanted, then you can retry again and again. In this example, I use Nano Banana Pro. You can also swap your model from P two or from Nano Banana two to P because in some cases, the Nano Banana Pro is far more realistic and much better than the Nano Banana two. After I tie, make modern mustard of this set in 2026 and have a look. This time, this is what we got. And finally, we have this taken screenshot. It make modern mustard of this, and this is what we got. Now, if you notice something, it swap my character, which is this one, this is also called the Jackie Chan in ten, and it swap the character to this gene. So the Nanoma two, which I noticed is make some unadditional changes that you don't want in your image, and it just make those changes on its no 14. Multi Element | Nano Banana 2: Welcome back, everyone. With the Nano Banana, you can also merge multiple elements. So as you can see, guys, if I click here, we have this abandoned car, alien dog. We have this rusty bike. We have this luxury couch, and we have this person. And as you can see, I have this prompt, and this is the result. So we can combine multiple images. Now, if you are wondering where I got this prompt, simply go to ato.google.com and you have to go to playground after you go to playground, I have tried this prompt. I have an image of abandoned car, rusty bicycle person, alien dog, and luxury couch. Give me five interesting ways to put them together in an image. So after I got the prompt, this is my first prompt. Now we have another example. We have this one. Now we have this result, and final, we got this one, we have this one. Now in this example, as you can see, guys, we have this person, this boy, we have myself. We have this one, this one, and we have this. And I type all of them standing on moon surface. This is the first variation and it's kind of weird. This is the second variate. And as you can see in this image, I don't have any leg shown in this image, so that is why it imagine my leg. I mean, this is weird. Besides if you use the same kind of art style, you can create something like this. I type all of these standing in All Street, doing Spider Man me pose, and this is the second variation of the same prom. Then I use my image and I use this vector style person and I say both doing handshake, and this is the second variation. And finally, I use these two and I say both holding each other here, punching each other. We have this second variation, and we have the final variation. So this way, you can generate consistency using multiple elements. 15. How to Generate Text-to-Speech: Welcome back, everyone. So now I'm going to show you how we can generate text to speech using AI studio for completely free. First, you need to go to chat and you have to select audio. As you can see, we have five options, and one of them is audio, so we can select the audio and we can select this Gemini 2.5 preview TT as soon as you select this one. Right here, we have the raw structure of our dialogues. This one is the style, how your speaker will speak the words. This is the style. This is the first dialogue. This is the second dialogue. This is our first speaker, and this is our second speaker. By the way, if you want to generate only single Speaker audio, then you can select the single speaker audio. And now we have the single speaker audio. And right here, you can select any voice that you want to play. By clicking on this play button, you can preview any voice. What do you want to explore? Once you select your voice, you can type any text that you want to convert that text into speech. Now I have typed this text. To convert this text into speech, you have to simply click on this Run button. Hey there. My name is Faison, and welcome to this course. As you can see, we have this style, so I'm going to change this style to Angry. And when you run this again, Hey there. My name is Faison, and welcome to this course. As you can see, it changed the tone of the voice. So now I'm going to type set style. Hey there. My name is Faison, and welcome to this course. Now let's change this to confuse. Hey there. My name is Pisan, and welcome to this course. Now, if you want to download this voice, if you want to download this text to speech, you can simply click on this player and you can select Download. And as you can see, now we have this WAV file. Now, if you want multi speaker audio, you can also do that by clicking on this multi speaker audio. This is our first speaker voice, and this is our second speaker voice. So I'm going to change my voice to this one, and for the second speaker, ready to make What sparks your interest? I'm going to select this one. Now, as you can see, it has already typed these dialogues, and we also have the style. So I'm going to run this. Hello. We're excited to show you our native speech capabilities where you can direct a voice, create realistic dialogue, and so much more. Edit these placeholders to get started. One thing that you keep in your mind, if you change the style of your speaker, it will affect both speaker, not the individual. So in this example, I change my style into bold and angry tone, and if I play the voice. Hello. We're excited to show you our native speech capabilities where you can direct the voice, create realistic dialogue so much more. Add these placeholders to get started. Now, sometimes you get the first speaker speaking only in this audio, so you have to refresh if you are getting this kind of error. Now, the easy part and the best part of using this Gemini TTS, you can add as many dialogue as you want. So for example, I'm going to add another dialogue, so I'm going to just click on this ad dialog and we can type anything. Now I'm going to add a dialog and now we have the second speaker. We can also delete our previous dialog. Now I have write these dialogues, and now I'm going to run this again. Hey, what's up? You going? I'm fine. Thanks for asking. So what's your plan for this weekend? Nothing. Just staying at home. Great. And that's it. This is how easy it is to add dialogues and add conversation between both speaker. And now we can also download this audio as well. So this is how we can use text to speech by using Gemini. 16. How to Do Prompt Engineering: Welcome back, everyone. So now I'm going to show you how we can do prompt Engineering in easy way. So first of all, go to labgooglelashflaool slash WISC and you can Inertol We have this interface. Go to astogogle.com. And we have to select chat. Once you select the chat, you have to select Gemini. Once you select the gemini, we have to type this simple prompt. Now we have to type this prompt, act as a prompt optimizer, enhance my prompts in cinematic style. And as you can see, guys, now we have to input our prompt. So first of all, I'm going to type glass of water and press Enter. The whisk is using the latest image and for model for image generation, and the best part of using the whisk is that you don't get any watermark. As you can see, this is a simple prompt result. I'm going to copy my prompt and I'm going to paste this prompt. And as you can see, guys, we got this precise prompt. I'm going to copy this prompt, and I'm going to paste it right here. So now you will notice what is the difference between the basic prompt and the enhanced prompt. Now, check out the result. Now if you notice we type cinematic style, that is why we are getting this cinematic style prompt. If you type, let's say I'm going to go to chat, and now I'm going to say, let's go to our previous prompt. I'm going to copy this one, press Control C, go to chat. I'm going to pay this right here by pressing the Control V. Now I'm going to remove the cinematic to movie L style. And now if we paste the prompt, let's open the image, copy the prompt, and now we can paste this prompt. Now check this out. I'm going to copy and I'm going to based this prompt right here. And now check out this result. So by doing this, you can enhance your prompt in any style. Now, if you don't want any style, you just want to enhance your prompt in like say, general manner, then you can also do that. And we have two formula. One is the formula, and second one is for enhanced prompt formula. So go to chat and I'm going to paste this formula. So now I have typed act as a prompt optimizer in this style. We have the main subject scene and motion, and I add a coma and I write write complete prompt after. So after we give him the prompt, it will work with this formula, and then it will write a prompt after this formula. So now we have to give him the prompt, so I'm going to copy my basic prompt and I'm going to paste this right here and I'm going to add a coma, and I'm going to write hand grabbing the glass. So I'm giving him some more information to work around with my prompt, so it can adjust my prompt according to this formula. Now let's run this and check this out. We have this basic prompt. Not only we have the prom, we also have some image as a reference. So now we can go to the whisk we can paste this prompt and now have a look. Check this out. Now we can also click here and we can also log the seed, but I don't think we have to do that for this prompt now we have another formula for enhanced prompt. So I'm going to click here and I'm going to copy this text, click here, and I'm going to paste this text. Now for enhanced prompt for the precise prompt, we have this formula. I'm going to paste my formula. First, we have the main subject, then we have the scene, we have the motion, and then we have the camera moment. After that, we have the aesthetic atmosphere, and that's it. I'm going to run this. Let's go to this. I'm going to copy my prompt. I'm going to paste my prompt right here and I'm going to add a coma. So now I have typed this prompt so that the gemini work around with my prompt according to this formula and write a precise prom. This is the camera moment, so I just type top you. After the camera moment, we have this aesthetic atmosphere. So I just type add cinematic lighting atmosphere. Now check this out. We have this prom, so I'm going to copy this prompt, and I'm going to paste my prompt right here. Now check out these results. So this is the basic result, and this is the precise result. Now you can work around with your prompt if you are not getting the result that you want. So for example, in this image, we ask for top view, so we did not get. So what you can do, you can go to define and you can type your camera moment or camera view. So as you can see, we do get a little bit different image than this one. So we did not get the top view, so I'm going to write again. So this time I type make top camera angle view of this image, and now have a look. We have this top camera angle view of this image. Now we have another way to enhance our prompt. So I'm going to copy my prompt. I'm going to create a new chat and I'm going to say enhance this prompt and I'm going to add this column, and I'm going to paste my prompt inside this and just run this. And as you can see, guys, the gemini will enhance your prompt. Now, after that, we just type enhance this prompt and we do get this enhanced prompt. So I'm going to copy this as a text, go to ISC. We can close this. We can paste the enhanced prompt. And now check out this result. This is the enhanced prompt result. Now, if you want some changes in your prompt, then you have to ask Gemini before you ask it to enhance your prompt. So let's say you want some cinematic lighting in your shot, you have to write that in your basic prompt, and then you have to ask the gemini to enhance your latest prompt. So the gemini will enhance your prompt according to your basic prompt. So in this case, we have this prompt. That is why we are getting this enhanced prompt. So if let's say have this basic prompt, you can see right here in this three word prompt, we don't have any context where the glass is placed, what inside the glass, et cetera. So if you ask the gemini to enhance this three word prompt, then the gemini will enhance this prompt according to his idea. So it will add points. I will add missing points in your story, and then it will create a new story according to his style, not your style. So if you don't have any contacts in your prom, then it will enhance your prom according to his style. So now let's see what the gemini come up with and now check this out. We do have glass of water, but the detail is filled by the gemini. So we don't fill any detail. The gemini automatically fill the missing points, the missing pieces of the puzzle according to his mind. So if you let's say ask the gemini glass of water placed on wooden table, light or let's say, sunlight directly hitting the left. And I'm going to also add a command. I'm going to type hart sunlight. I'm going to also add dart room. And now, as you can see, in this prompt, we have so much context. Glass is placed on wooden table. We have sunlight directly hitting the glass from the left side. We have harsh sunlight, and we have dark room. Now, this time, if you ask the gemini to enhance your prompt, then it will come up with an enhanced prompt according to your missing pieces. So now I'm going to paste press Control A, press Control V to paste your prompt. If you press Control A, it will select all the texts, and now have a wooden table, glass of water. The heart sunlight is hitting from the left side, and we do have the dark room. So this is what I was talking about. So keep this in mind before you ask the gemini to enhance your prompt. And these are some easy method to have a precise prom from a basic structure. And this is the easy way to do prompt engineering using the Gemini. 17. Steal Any Prompt: Welcome back, everyone. So now I'm going to show you the most important part in AI content creation is how we can copy other images prompt, how we can extract prompt from an image. To do that, you have to go to is click here, and in the subject, you have to upload the image that you want to extract prompt from. Once you upload your image right here, we can click here onto this icon. As you can see, we got the prompt to generate these kind of images. Now I'm going to press Control A, press Control C to copy my prompt, or you can right click, you can copy your prompt from here. Close this, we can delete the subject, and right here, we have to paste this prompt. Now if I click here to generate my images, you can see we have similar result according to our reference. This way, you can copy any image from the Internet. Now to get these kind of reference images, you have to go to Pintrest and you can type AI animation or vector animation. To get these kind of images, you can type ai prompt in pintrus.com. Now why I'm telling you this is because I'm going to show you some interesting part of AI content creation. So I'm going to press Control A, and I'm going to backspace to delete my prompt. I'm going to click here and I'm going to use my subject, and this is my subject. By the way, to create this kind of subject, what I did, I just use astudio.google.com. I upload my image by using the nano Banana model. As you can see, this is myself, and I just ask nanobnana make me wear a ski mask with yellow and blue outline. And as you can see, guys, this is what we got. So at the end, I just make mask full yellow, and this is the final result. And this is the subject that I am using in is. Now I'm going to make my subject in this style. So what we can do, we can download this style. We can upload this style. And now we can convert the subject into this vector style. Now check this out. Now we are converting the subject into this style. You can also make your subject do anything. With this style, I just ask the make subject look straight into camera and the subject is looking straight into camera. So you can ask it to make your subject do anything that you want. Now, why the subject is not in this style, if you are getting this kind of result, you can simply in the style click here, we can copy the style, and we can also plus shift plus nter so it don't generate your image, and we can paste the style. We can add a comma right here. Now click here to generate your image, and now have a look. We have the similar style and the subject is looking straight into camera. Now we can copy any style that you want to. So I'm going to delete my style. If you remember in print terss as you can see, guys, we have these kind of images, so you can copy any image style that you want. Now I have uploaded this style. Let's click here. I'm going to press Control A, press Control C to copy this prompt. I'm going to click here, press Control A and Control V to paste my prompt. Now, after you paste your prompt, after you use this style image prom, make sure that it don't contain any subject detail or description because if your prompt contain any subject description, then it will generate a brand new subject image, not with your own customizable subject. Now I'm going to delete my style because we don't need that because we have the style prompt right here, so we can generate images without any style and check this out. This time, I'm going to use this reference image. Let's click A. I'm going to press Control A, press Control C to copy the prompt. Let's delete the style. Double click, and we can paste the prompt right here. Now check this out. Now you don't have to describe any changes in this prompt. You can also go to refine and you can ask you to make some changes like. So I just make him look straight into camera and make him put his hand in the pocket and check this out. Now let's get out from this refine mode. Now this time, I'm going to copy this style, so I'm going to click here. I'm going to press Control A, press Control C, and press Control V right here. Now if you keep the style, then you will get different result. And if you remove your style image, then you will get different results. So in my opinion, the best way to generate images is to keep your style, then generate image. And after you generate your images, remove the style image, keep the prompt, and then generate again. This way, you will get different variation. So now we have this style. So I'm going to press Control A, copy this style, click A, press Control A, Control V to paste my style right here. Time, I'm going to generate my image with the style image and with the prompt, and then we can generate without any style image. Now let's delete the style and click here to generate images. By the way, you can generate multiple images. You don't have to weight your previous image to generate. You can generate multiple images. These two are generated with a style image and with a prompt. And these two are generated without the style image. As you can see, guys, these two are slightly different than each other. So the best way is to keep your style image, generate images, and then remove the style image and then generate images. And in both example, keep your prompt, keep your style prompt. Now in this example, as you can see, guys, my prompt contain the subject description. So we have to delete the subject description because if we don't delete the subject description, then it will generate images with a brand new subject, not with our own subject. Now, without the subject description, we can generate these images with a style image reference. Now if you do get this error with a style image reference, then you can delete your style reference, and with the prompt, you can generate your images and boom, check this out. So this is how we can copy and image and style from the Internet. 18. JSON Prompt Engineering: Welcome back, everyone. Now, I'm going to show you how we can write JSNPmpt. A Jason prompt is basically a machine language so that the AI can better understand what we are trying to generate. It is a little bit better than the simple English. Create a new chat by simply clicking on this chat button slash Gemini, and you can paste this prompt. You can use this prompt. Give me 50 ideas for generating images in Imagen three. You can also type Imagen four. If you want to now I got this crazy idea. Now to write JSON prompt, we can simply use this simple prompt. Click here, create a new chat, go to Gemini. Click here and you can paste this prompt. Act as JSNPmpt writer. Now, make sure you write JSON in capital from simple English into JSNPmpt structure prompt. And now let's click here. Now as you can see, guys, we got the JSON prompt structure. This is basically how JSON prompt is look like. Now go back to the history or you can right click and open this in a new tab. Now open your project of Gemini where you type 50 ideas for generating images, and you can copy any prom that you like. So for example, in this case, I'm going to copy this prompt. Let's copy this prompt. Now go to WISC and you can paste your prompt without any of the subject scene or the style. And the WISC is also using the Gemini for model. Now let's generate the image with a simple prompt. Go to your act as a JSON prompt writer project and paste this prompt. Once you run this, it will convert this prompt into a JSON prompt. Now we can basically copy this from here, go to is, and we can paste this prompt. Now, this is the result of the simple English, as you can see, and this is the result of the JSON prompt. As you can tell, both of these images look different than each other. With the same exact prompt, we got two different results. Isn't this amazing? Have used this method, and I have generated several images. And let me share with you my example so that you can better understand what is the JSON prompt. Now we have this example. As you can see, this is my simple English prompt, and this is the result that we got from isc and with exactly same prompt, I just convert it into a JSON prompt with the same method that I just share with you. And I mean checkout the result. The lighting is changed. The scene vibe is completely changed. I mean, this looks like illustration, and this one looks like a three kind of a scene. Now with this prompt, this is the result that we got, and with the JSON prompt, check out the result. This is my JSON prom. Now with this prom, this is the simple English result, and this is the JSON prompt result with the exact same prompt. Now we have this image. This is my prom. And as you can see, guys, we have error in our image. And this is not a single image. We have two images with the same error. But with the JSON prompt, it also fix the error, and it also make the scene a little bit more scarier than the previous. I mean, checkout. This is simple English, and this is JSNPmpt. Jason Prom result is looking more realistic than the simple English. We have this prompt. I mean, check out the result. We have the same error, and with the Jason prompt, I mean checkout result. By the way, this is my prompt. For this scene, this is my Jason prompt, and this is my simple prom. We have this example where we have this prompt. We have this scene with simple English, and with the JSON prompt, I mean checkout result. Jason Brohm perform a little bit better than the simple English. Now we have this example. As you can see, by using the simple English word, we are conveying the message to AI, and we got this image. But if you look closely, as you can see, guys, JSON Prom result is looking more realistic, and it also understand my prom better than the simple English. Now, if you look closely to my prompt, I have typed a shaky handheld shot of a flickering street light. Camera struggling to focus faint silhouette moving in the distance. And can you tell we did not get anything from the simple English, but with the JSON prompt, I mean check out the result. Camera is struggling to focus. We have shaky handheld camera. It also organize your prompt so not only the AI, but you can also understand it better than simple English prompt. We have this example where we have this simple English. The person is running. With the JSON prompt, I mean checkout. We have some different variation than the simple English. Now we have this example. This is the simple English result of this anime, and this is the JSON prompt result. In this case, I think the simple English is performed better than the JSON prompt. Now we have this example where we have this long prompt, and with the JSON prom, this is what we got. In this case, too, I think I really like the simple English result. Now we have this prompt. This is my simple English prompt result, and this is the JSON prompt result. I mean checkout. The JSON prompt make it more dramatic than simple English. We have this example where I use this prompt. This is simple English, and this is JSON prompt. I mean, check out the result, both are different than each other. We have this prompt, and this is my simple English result, and this is my Jason prompt result. Jason Prompt make it look more realistic and believable than simple English. I mean, this scene is looking like it was shooting from a camera holding by a person in a helicopter. Now I got this image. This is my prompt, and this is the simple English result. And this is my JSON prompt result. I mean, check out the quality and the result. It also changed the camera angle, and it also improve the image than the simple English. Now we have this prompt, and this is my simple English result, and this is my JSON prompt result. Now, in this case, Jason Brom just make the image shine a little bit more than the simple English. In simple English, as you can see, we got this classic style, but in Jason Brom, we got a kind of a reboot result from a son Brohm. It just polish the image a little bit than the simple English. Now with simple English, this is my simple English prom. And as you can see, guys, we have a better result in Jason Prom than the simple English. Now we have this example. This is my prompt, and this is simple English. And with the same exact prompt, just convert it into the JSON prom. This is the result that we got. Now we have this example. This is my prompt. Now if we compare these images, this is a simple English. This is the JSON prompt. You can probably tell that the JSON prompt is performing better than simple English, a little bit, not too much, but a little bit. Now we have the final result. I use this prom to generate this image. As you can see, guys, we have the simple English result. But in Jason Brom, it looks more realistic than the simple English. I mean, check out the lighting, the camera angle, the depth of field, even in the image is more realistic than the simple English. Have a look. So this is the easy method to write JSON Pm. 19. What Image Generator You Should Use? | Imagen 3 VS Imagen 4: Welcome back, everyone. So in astudio.google.com, if you go to chat and if you select images, right now, we have two major model. One is Imagen four, and the other one is imagen three, which is the previous model. So which one is the best for image generation? If you think that imagen four is the best because this one is the latest model, then you are wrong. Let me show you some examples so that you can better understand which model to use for generating specific images. Now, I use this prompt and by using Image and four, I have generated this image. Now, if you look at the model, all the images shows the image and four model. Now the reason why all these images shows Image and four model is because we have selected image and four model. If you have changed your model to Image three, now look at the model, all the images now have changed their name to imagen three. So this is just a display model name, not the images generated by the specific model. Now with this prom, this is the imagen four model result. And as you can see, guys, the images are in high resolution, and this one is generated by imagen three model. Now I have generated more example. This one is from the imagen four ultra. Check out the result. And this is again by Imagent three model. Now I want you to look closely in the Imagen three model, as you can see, this one look like straight come out from a movie scene, while the Image and four model looks straight come out from a video game. The reason behind why we are getting this kind of result from the latest model, while the previous model nailed the performers of my prompt is because this model is used to generate realistic images. While the Image three model is better for generating imagination if you have some crazy things going on in your prompt, then you probably should use Image and three model. You want to generate realistic high resolution images, then you should use Imagen four model. Because if you switch your model to Imagen four, you can generate your output up to two K four completely free. Now we have another example where we have this three D scene going on. This one is from imagen four model. As you can see, guys, everything is look pretty sharp and everything looks amazing, like a three D scene. While the Imagen three model also generate crazy scene, but as you can see, guys, we have some plastic things going on in the model. I have generated this image again with the imagen four model and check out the result. This look like straight come out from a movie scene, while this one look straight come out from a three software. Now we have this prom and this one is generated by Imagen three. And with the same prom, this is the imagentF model result. As you can see, the quality is in high resolution with two K quality. We have this image. Now we have another example with this simple prompt with the imagen form model with two K resolution. As you can see, guys, we got this crazy result, and with the same prompt, we have imagen three model result. If you notice the imagen three model is in low resolution, while the imagen four model is high resolution plus it generates realistic image. Here are some examples. With Imagen four model, we got these kind of game like images with this prompt. While the Imagen three model, as you and see generate more realistic images than imagen four model. Check out the result. Have another example with this prompt. This is the image and four model. As you can see, everything look like straight from game. But in this example, we kind of get a game image, but it look more realistic than this previous image. Now we have this example with this prompt. As you can see, guys, we got this result from image and three model. And for some reason, image and four model generate this kind of image. I think it misunderstood the ducking and it literally generates a duck face person. Now we have this example. This is my prompt, and this is Image and three model. As you can see, it generates a pretty realistic result. Now with the same prompt for some reason, imagen four model generate this kind of image. I mean, the image is in high resolution, but check out the result and tell me which one is looking great, imagen three or imagen four. Now we have this example with this prompt. We have imagen three model result. As you can see, we have this pretty crazy result. With the same exact prompt, this is what we got. I mean, check out the quality and everything look crystal clear. Now with this prompt this is the result of imagen four model. As you can see, guys, this is some crazy result in high resolution. And with the same exact prompt, this is what we got. As I told you earlier, if you have some crazy things going on, use Imagen three. If you look closely, we have depth of field. We have crazy motion. While in this scene, we have a game kind of image result. You can compare by yourself and tell me which one is the best for these kind of crazy ideas. By the way, this is my prompt. Now with this prompt, if I open my image, as you can see, guys, we got this result and the result is not bad. And with the exact same prom, this is the image and three model. I mean, both model generate images like it's straight come out from a game, but I think in this case, the image and four model nailed the performance but I asked for a personal recording, so the imagen four did not deliver what I asked him to deliver, but the imagen three model understood my prompt, and this is what I got. Now, with this prompt, as you can see, guys, this is the result that we got. And with the exact same prom, this is the image and for model. And as I told you earlier, if you have realistic prompt ideas in your mind, then you can use imagen for model for realistic ideas. I mean check out the cinematic of this image. Now with this prompt, this is the result that we got, and I really like the performance of this image by Imagen four and with the exact same prompt, this is what we got. In this case, as I told you earlier four realistic images use Image and four. That is why in this case, we have this game image because we used Imagen three. Now with this prompt, we got this result. As you can see, guys, even we have some crazy imagination, the image and four deliver the best images that it can. While the Imagen three model deliver this kind of image, have some cinematic going on in this imagen three. Now with this prompt, we got this result. And this is some game kind of image. And with the exact same prompt, this is what we got. And you can tell which one is the best by looking at the image. I mean, the quality is in high resolution, but I did not like the result of image and four. Now with this result, I'm really surprised because as you can see, guys, I use this simple prompt and I did not put any detail in my prompt. And this is what I got from Imagen four model. I mean, check out the result. It looked like it was hand drawn by some person. And with the exact same prompt, this is what we got. I mean, I really like the both result. Imagen three model delivered the best, but the imagen four model in my case, is the best if you want to generate CD animation. Anime, if you want to generate realistic images, et cetera, you can use imagen four model. But where the imagen three shine is if you have some crazy imagination in your prompt, if you have some crazy stuff going on in your prompt, then you should probably use the Image three model. The example, where we have this pixel style boy with rough hair standing, check out the result. We have some plastic key results like it was rendered by a blender three. But with the exact same prompt, if you check out the result, I mean, you probably can tell that this is the best output from Image and four. So for three D animation, you should use Image and four model as well. Now with this prompt, we got this result. I mean, check out the quality, the lighting, everything looks amazing. And with the exact same prom, this is imagen three result. I mean, in this case, both, I think it is a tie for both, but you can collect both of these images. If you like the color of imagen four, you can make Imagen three model image look exactly like this one by correcting your image. So it is not about collect correcting. It's about the image performers. So in my case, I think both are good for these kind of images. We have this example where we have empty college in a desert, and I mean imagen three model just generate a drone kind of image, but the image and four model delivered the best and it generates this kind of image. Now we have flat vector style group of boys, and this is what I got from image and four model. And with the exact same prom, this is image and three model result. I really like the vector style of this image as well, but in imagen three model, you can see, guys, we have one, two, three person duplicate. We have one, two, these two are duplicate, these two also duplicate, and this one is different. If you look at the expression, only three are different expression, and these are some duplicates. But the imagen four model did not do this kind of stuff, and it delivered a completely different image, a completely different person than each other than image three model. Now we have this example where I just type car drifting, and this is what I got from imagen three, this is not bad. I think this is really great that it come from imagen three model with this quality. And with the exact same prompt, I mean, check out the result. Isn't this amazing? It looks like someone is shooting with their camera. Now, with this prompt, this is the result that we got from Imagen three model. And this is the result of image and four model. Now we have this prompt, and this is from image and four model. I mean, check out the quality and image performance. With the exact same prom, this is imagen three model. I mean, in this case, image four model perform a little bit better than imagen three model. Now we have final example where we have this prompt and we got this result. As you can see, guys, we have some crazy stuff going on in the prom. That is why we are getting this kind of crazy image. While if you look at the imagen four model result, it just delivered this image, but I really like the quality, the image performance of Imagen three model. Now if you want to get the best result from this model, then you should use this method. Step one is copy your prompt and paste it in Imagen three model, then generate your second image with Imagen four model. Step two, once you got two images, you can compare them together, and then you can find out which one is the best looking image than the other image. Step three is to download that image and you can use that image in your future projects. By using this technique, every time, whether your prompt is good or not, you will get the best result from these two model. 20. Use Cases of Nano Banana: Welcome back, everyone. So now I will show you the use cases of nano Banana. First one is if you want to remove the person background, you can remove it easily. You can ask remove the background. And as you can see, guys, we got this perfect person without any background. I mean, check out the adds. We don't have any rough add. Not only you can ask it to remove the background, you can ask to put this person onto another background. So I just ask place person onto another background. You have this user, it means this is your prompt, and if this is, say, the model, then it is the AI prompt. So after my prompt, the AI nanoba ask me to where I want my person. I want to place it in a forest or I want to put this person in a CD square. So I just type forest, and as you can see, guys, we have this person in a forest, and I mean check out the quality and check out the person lighting is perfectly matching with the background. Can also remove anything from your image. So let's say you want to remove this plant. So I just ask remove plant. And as you can see, guys, it perfectly removed the plant from this image. Not only this, you can also make this room empty by asking to make room completely empty and check out the result. For example, if you have the thumbnail and you want to keep the design, but you want to remove the person from that background, you can also do that by asking remove the person. Sometimes you will get this kind of results, so you can rephrase your word. So I just type remove person from image and check this out. It perfectly removed the person from this. Now I can place any subject in front of this thumbnail. Now we have this redesign. Now you can also mess around with the color. So I have this vector with this blue color and I just change background color to red and check up. Now, for example, if you have this image, you can also change any color that you want. I just change was color to purple. Check this out. Now, for example, if you have some real life images like I have. So in this image, I want to make the apple red. So I just make Apple look red and check out this. Now, I want you to look closely the apple in the original image and the apple that AI generated is completely different. If you look at the apple, these are completely different apple. So the model, the nano banana completely generated new apples. So in this type of scenario, you can type this prompt. So don't have to upload your image every single time you want to ask for any changes, you can just right click, click on the copy image, click here, and you can paste your image. And now you can ask for any changes that you want in your image. Now I just paste my image, and I just changed the color of the apple to red. Just change the color. Don't make new apple. And as you can see, this time we got the pretty close result to the original image. You can also mess around with the adjustment colors and brightness controls, et cetera. So I use this image, and I ask adjust the brightness, increase it by 10%. And as you can see, guys, we have 10% increased brightness in this image. If you look closely, we have 10% brightness increased in this image. I just ask adjust contrast, increase it by 20% and check this out. By the way, we did not get any changes in this image. It just make this image a little brighter. It did not increase the control by 20%. So sometimes it just give you this type of result. Now, this time, I ask it to adjust vibraus, increase it by 50%, and we got this. As you can see, this is a pretty bright result. Not only the adjustment of the light, you can also remove the noise in this image, as you can see, we got this pretty noisy image right here. So what I did, I just ask it to remove the noise and check out the improved results. You can also color grade your images. So I just ask add matrix, color grade. Have a look. Now, this time I ask add teal and orange color grade and check this s. Now, if you want to color grade your image, you have to copy and paste your image right here every single time you want to ask for a new color grade. Because if you are color grading and you ask you to add teal and orange onto this matrix look, then the result will be different than this one. So this is the teal and orange result. Now this time I ask to add cinematic color grade and check this. We have this cinematic color. Not only this, if you want to restore the color of your images, you can also do that very easily. So in this image, I just use photob.com, which is the free Photoshop, and I just lower my saturation to about 0%. And I just ask NanobnanaRstore the color of this image and have if I show you my original image, this is my original image, and this is the nano banana restore color image. And the result is pretty remarkable. Now in this example, I just use this anime image, and I just ask restore the color of this image and have a look. Now, if I show you the original image, this is my original image, and this is restored by nanobnana and I really like both results. The nano banana did a pretty good job at restoring the color of any image. With nano banana, you can do anything that you want. So I have this image, and I just as nano banana person taking selfie with iPhone, and as you can see, guys, we have this person taking selfie. Not only this, you can also mess around with the camera angle. So I just type make it a white angle and check. We have a white angle result of this image. I again ask you to make it wide angle of this image and as you can see, guys, we got this image. Now, if you have this image and you type a prompt, then it will make changes according to the final image. But if you want to make changes to the original image, then you have to copy and paste your image right here or you can add another image from here or you can take a photo or you can use your drive to upload the image, and then you can ask it to make changes to your picture. Now this time, I ask you to make it Fish Iview and check this out. We got this amazing fish Iview now I just use my original image, give me close up angle of this image and check this out. We have the close up of this image. Not only this, if you want to type any text, then you can do that very easily. So I just use this image, analyzes type, add a text behind subject, and in this column and type, Vo 3.1, make it look like it is written on the wall with spray, and check this out. We have this VO 3.1 written on the wall. As you can see, we got a little bit error. So what I did, instead of asking it to make it correct, I just ask Make text three D, and this time, as you can see, guys, we have three D text placed behind the subject. Now this time, I just type now write it on person shirt like it's a monogram and remove it from the wall. Now you have to define every tiny detail if you want to get a perfect result and check this out. We have this video written on this person. Chart. If you want to stylize your images, you can also do that very easily. So as you can see, guys, I have this image and I have this reference. By the way, if you want to find the reference, you can go to the Pintress and you can type image effect, and you will get these kind of images. Now I use this problem. Use reference style of the right image and turn the right image like that and check this out. We got this result. So it mimic the effects and apply this to your image. Then I use this effect and it is type, now use this reference and have sometimes you will get this kind of result, but don't worry, you can rephrase your word to make it look better. Now I just type just use the style and have a look. It just replace my face onto this person. Now I just type now use the reference and apply that and check this out. Now, the more you ask for any changes, the better it will get every single time. Now I just use this reference and I just type use this reference and have a look. By the way, I did not use my image because in this chat, we are starting with the style, so that is why the AI is understanding that I want to copy the Image style. So that is why it just follow my prompt. So make sure you create a new chat and you apply your style onto that chat. Don't mix your work in one single chat. Create each chat for each single work that you want to do. Now this time I just use this reference and I got this result. So I just say refine it. I just add outer glow onto my image. Now I just make it a bit darker, also add more glow and refine it more and have a look. So you get the idea. Now I use Joker image, and I just apply Joker makeup on to the right image and then I use the single image without my image. And I type, use this image as a reference and apply the grading of this to previous subject standing image and check this out and have a now I use this reference Nags type, this prompt and have a look. I did not get any good results from this prompt. So I just again use these two images, and I rephrase my prompt energ type, use the reference on the right image and apply the image effect and lighting color to the left image. Just use the style and now have a look. Isn't this pretty crazy? Now, if you are working for clothe branding, then you can also do nanobanana. So I just use this teen image, NDS type, make person wear this jean, but I did not get any result out of it. So I just type person wearing black jean and this time, I use new image and check this out. We did not get anything. So this time, I specifically type use Jean and make the person standing wear it and now have a look. Now I just use this jacket ANDs type person wears this jacket and check this out. I use the set I type make person with set. Now, if you are not getting the result that you want, then you try to swab the item and try to rephrase your prompt as well. So this time, I just type this prompt and have a look. You can also do this type of work like we have this open AI and we have this font and I just type change at dipty logo to Google logo and change open AI to Google and have a look. It perfectly matches the blur and it also matches the font of this open AI. Now if you are a graphic designer, you want to design a business card, then you can also do that. You can create a business card for this logo for video editing insta FB, UW, et cetera. You can use your logo, and you can create a brand new card from scratch. You can design your logo very easily. So I just use my logo and I just te make this metallic logo. We got this, and I just make it more realistic and rusty, adjust this texture and mate the style. Check this apply volcanic style and have a look. Then I use my logo, which is Dantrol Antes type now applies same style to this logo and have a look. Then I use the Kling AI nets type this prompt, but I did not get anything, so I just use my image again, Annetsty make this water texture, and adjust the color. Then I use the image again at, make water texture to this logo, and now have a look. Then a type, make it more clear and brighter and check this out. Then a type, refine it more look. Then I use Capcud logo and IS make it file logo. Then I type remove white stroke from this logo because we got hit stroke and it perfectly removed the white stroke. Then I use King logo again and adds type now apply previous style to this logo. Then I use image again, alt make it file logo. If you are not getting the result that you want, then try to rephrase your words, then I use my logo again, and at make this logo file. Have a look. Then I use the Kinga logo again and just type now make this logo here and have. Then I ask you to make it s. Then I have to create a new logo. So I just type, create flat logo of FA, we got this logo. Then I use my reference style. So for again, the reference style, you have to go to Pintres and type logo. You will get this type of I do style make FA logo look like the right image style and have a look. So you can use any style and you can make your logo look like that style. So nanobnana just make the logo designer work so easy that even the amateur can do logo designing. Now, if you are a graphic designer, then you can also use nano baanana for your have this example, another stype change the text to eat the billboard and have a look. I mean, it matches the text, and it also matches the font, every single detail. I matches with the original billboard. Next, I just make billboard logo upside down, also change the tag line to I am heating it, also the all white meat chicken to vegetarian burger. So this is all white meat. I want to change it to vegetarian burger and now have a look. We have vegetarian burger, but we did not get the logo upside down, but we have Im heating. So just type change McDonald logo to upside down, have a look. Then I'd ask you to change border on billboard to a bunch of salad. Now I meant to type the burger, so I just accidentally type butter, but understand it anyway. Next, I ask you to change color palette of this billboard to blue and orange and have a look. Now, this is pretty useful for graphic designer because as you can see, guys, if I have to use the Photochop then it will take me hours to create this type of billboard, this type of change. Then I ask you to now change to green and blue and have a look. Now, if you are working for a brand and you want to design their poster, you can use nano Banana. So I just use my FA logo and I just use this style. By the way you can find this style from the printers, you can type logo. You will get this type of style as well. And I just make FA logo ply style of the refers image. So we got this logo. So I use this logo, and I just type, create a sleek juice box with this logo. We got this I use this again and this type now, create a orange flavor juice with this logo. We completely created this new juice box from scratch. Then another type now make for apple flavor and heel. So we can create multiple flavor. Instead of using the Photoshop for mockup designing, we can use nanobnana. Now we are going to create a commercial out of this juice box. So I just type now make this juice box land or ground with water splash and falling bunch of apple from the above and have a look. Then I use my image again, another time make the juice underwater with a bunch of apple slices floating inside the ocean. Have a look. Then I use my juice box again, another time make this juice box flavor mango, have a look. Then I use this reference style, which by the way, you can find this from Printers. Now, after we get this mango flavor juice, I use this reference image, which by the way, you can find it from the print. Type replace the person holding this glass with my juice. I want to replace these with my juice box as well, and I want to replace the text drink it and fill it. This is my prompt. So as you can see, guys, we got this replacement. We have the text, but the person is still holding the beer. So what I did, replace person holding beer glass to choose and have a look. Then I say remove two boxes at the right bottom, and it also remove these boxes from here. Then a type remove all the text from the image. Now we got this clean image. Now after that, I use my completely new reference no sty replace the can with Jose box and have a look. Then I use replace background with explosion and apple slices are flying away and check this out. We have explosion, apple slices, flying away. Then I use this meat type had water surrounding the person with apple slices inside it. So I just ask it to water in a spiral surrounding person, have a look. You can ask it to change that you want. Now, after this, I want to change the color to yellow, so we are getting these kind of changes. Now, this time, I just asked to add light yellow water in a spire surrounding person, don't cover his face, so I don't want to cover the face of my person, so that is why I type this. Then I say remove that cover person face, still did not get anything. So what I did, I just created a new chat. In this new chat, what I did, I use my image. And type as Spital light yellow water splashes surrounding person body. So this time, I just type person body. That is why it did not cover the pace. Then I say add Spitalight yellow water surrounding person body. We have this again. I type cover only body. We are still getting the same image. Then a type removes from the head and face. Now take this out. So you have to be specific what you want to change in your image because the AI only staand your word, don't understand your thinking. As you can see, guys, this way, you will get a brand poster for your product. Now, if you have a data, you want to represent it with visualization with graph, for example, you can create image of a graph. We got the image of a graph, then you can type your data, which I copy from Google. This is data by the way. Then I type make data visualization and for graphics of this data. And now have look, we have this graph. Then I type, refine the graph. We have this refine graph. I use a reference graph and a type convert the style to write graph and have a look. We easily created this graph style. Then I type now generate another version of this graph. We have another version. Then I type now make it a pie chart. We have this Pie chart. So you can easily visualize your data as well. You can merge the multiple elements into one single image. So we have this element, and I type person in a helmet wearing suit and shoes and leaning on the Bugatti in a beach and besides a superbike is part have this element and have a look, we have this spinal image. We have this jacket. We have the shoes person is wearing. Now this time, I just use my person and a robot and this couch I type two robots sitting on a luxury couch. Have a look. Then I use these elements and type make this item place on a brown computer table, top view angle, have a look. A type now shows left side camera view. We did not get anything. Then I use my image analyst type, show side camera view of this image and have a look. Then I use my image again, Analgs type, show front camera view of this image. We have this view. Then I use these elements, Nate make this item placed on a brown computer table. Then I use this element and check this out. This time we got this result. One of the most popular use cases of nanobanana is to make your character consistent. So I use my character and I use this pose as a reference. By the way, this is a CF pose. You can find it from Google, a type, change the pose of the boy and match it with the second image, which is this one and have a look. Then I use this pause and now show me this pause with then I use this pose and type now match pose with this. So the AI is using this image and that is why we are getting this result. So I just copy my image. This time I use my brand new image and type boy is doing the Cf pose of first image, which is this one and have a look. Then I use this image and a type now boy is doing this pose, the boy is doing the same exact. Then I use this pose and type now, boy is doing this pose, have a look. Then I use my image and type make boy sad crying. Have a look. The boy is sad and cry I use this image type Make Boy happy jumping midair. Then I ask you to change background to school, have a look. So you can make your character consistent in every single frame. Not only the animation, if you have some real life example, if you have some movie style, so this time, I have Pacific Rim style battle. So I just type this prompt. You can read it, you can pause the lecture and read it, and we got this image. So I just type low angle cinematic shot of this prompt and have a look. We have the low angle view of this shot. Type over the shoulder view from the Mac POV and have a look. So by doing this, you can change your camera angle and you can view your scene from every single camera angle. You can also make two character fight or interact with each other, like I use these two and type show me both our fight in ocean. We have these two in ocean. Then a type, show me both fighting, punching each other. We got now, there are some other use cases for nano banana. Like if you want to do motion graphics, you can also do that with nano banana. But these are the most useful use cases for nanobanana. 21. Introduction | Google Gemini 3: Welcome everyone. So now I'm going to show you how we can access to the brand new Gemini three. Simply go to astudio.google.com. Once you go to this link, we have this interface. You can simply select Gemini three or we can go to the playground. We can go to Gemini. We can select Gemini three. Now, once you select Gemini three, P preview, as you can see, we got these settings as well. We can also change the level of thinking from high to low. You can tweak the temperature as well. You want to access to the previous model, you can simply select Gemini 2.5 pro. This is the previous model. If you want to access Nano Brenna, you can click here. Now let's select Gemini three. Let's close this, and let's test this out. So now I'm going to type the symbol prompt, write a cartoon script for fire scene, a boy finding missing fish. Let's run this. And this is the new part of Gemini three. As you can see, guys, right here, he is thinking, so we can see what he is doing. And after the thoughts end, he is writing the full prompt. So you can collapse and you can expand the thoughts of Gemini. After this, as you can see, we have the final prompt. We have five scene prompt. Now we have another way to access two Gemini three. Simply go to gemigole.com. Once you go to this link, you have to click here and you have to select this thinking. Once you select this, you have to type your prompt and you can press in. Like the AI Studio, as you can see, guys, he is thinking, he is analyzing, defining, tracing the glue, et cetera. Now after that, we have the final prompt. Now you can also expand the thought by clicking onto this drop down arrow, and as you can see, guys, this is what he was thinking. This is how you can access to the Gemini three. 22. Build Simple Game | Google Gemini 3: Welcome back, everyone. So now I'm going to show you how we can develop simple apps. So to develop your app, you have to simply go to tools and you have to select Canvas. Now I'm going to type a simple prompt. Now as you can see, I have this prompt, create a colorful snake game with interesting UI. So I'm going to type this in a per case. Now, press center, and as you can see, guys, the gemini is developing the app. Right now he's thinking, he's considering the game mechanics. Now he is exploring visual, refining UI overlay. He's designing the core game. Now he's optimizing the grid rendering. Now as you can see, guys, finally, we have the game. So this is the code of our game, and this is the game. Now the best part about Gemini three is so good at making apps. Now as you can see, guys, we have this interesting star menu, so I'm going to click on the start, and as you can see, we have the zero score. So I'm going to control the snake with the arrow key, and as you can see, guys, the game is running accurately. And we also have the score updating as I am eating these dots and check this out. Now, if you want to fix the error in your game, if you have any error, you can also fix that. If you want to make some changes, you can also make those changes. So there are two ways to make changes in your app or game. First, you can type your changes. So I have typed camera follow Snake. Let's run this. And as you can see, guys, he is adjusting the camera view, expanding the word size, revisiting the camera logic, updating the game world. Implementing camera controls, enhancing world aesthetics, and I think he's done thinking. So let's collapse this. And as you can see, guys, he is updating these changes into this app. Now, after the gemini make those changes, the game automatically refresh. So now I'm going to start this game. So as you can see, guys, we got this error, so I'm going to type. So I've typed camera stuck fix error. Enter. The Gemini also is very good at fixing the error in your apps or game et cetera. Now, the Gemini fix the game and he is updating these changes into our game. And as you can see, guys, this is a long list of error fixing. So this is the best part you can view the thoughts of the gemini. Now let's run this game again, and as you can see, guys, the camera follow the snake this time. Let's scratch this and the game is working fine. Now we have this feature called select and ask, we can click here. We can select anything in the game or app, and we can make some changes. So I'm going to type, change the game over color to blue, and now I'm going to press Inter. So we can also edit the app or game by using this feature. Now let's view the changes. So as you can see, guys, we got some error. So let's start. And as you can see, we got some error. So if you have this message, you can simply click onto this Fix error, and it will fix the error that is stopping your apps to run. So after it fix your error, it will tell you what was causing the problem. Now, hopefully this time we can run this game. Let's crash the snake, and as you can see, guys, the game over color is now blue. So this is how you can use Gemini to develop simple game. 23. Build Simple App | Google Gemini 3: Welcome back, everyone. So we develop a simple Snake game. This time we are going to develop simple apps. So go to gemnigogle.com, click onto tools Canvas and type prompt to build your app. So I type this simple prompt. Build a PDF generator, ask user to input text and make it into PDF. Let's also type File, and let's run. So make sure you have selected the Gemini three by clicking onto this thinking. Now, within 60 seconds, it developed the app. So let's see if the app worked correctly or not. So now, as you can see, guys, we have to type the document title and main content. And I want you guys to look that this is a refine app. So let's type this, and I'm going to copy this, paste this, let's say download PDF, let's see if this work or not. And I mean this is crazy. Isn't this beautiful? You can generate any app that you want. Example, you can also build an app that convert your texts into uppercase. So I'm going to type a simple prompt. So I type this simple prom. Build app that chain text into upper case sensitive, lowercase and correct grammar. Now, if you are developing app, the best practice is to create a new chat every time you generate a different app. So I am generating the second app into one single chat. So this is not a professional behavior. If you professionally generate apps, now, as you can see, guys, it merges these changes into one app. So I want this as a separate app. That is the reason why I'm telling you to create a new chat for each app. Now we can also copy this, click here, click here, create a new chat. We can go to Canvas Paste disease press Enter. Now, this time, we are going to get a standalone app of these features. Now we got this app. Now I'm going to copy this text, and I'm going to paste my text here. Let's change this to uppercase. It is working. Change this to lowercase, title, sentence. By the way, we also have the words, so we have 68 words. We also have 480 characters. Now we can also fix the grammar, so I'm going to delete this text and let's focus onto this. Let's say I shall build a react application, so I'm going to change some spelling. Let's see if the app is working or not. Let's adds and let's add. Now I'm going to select my text, correct grammar. And as you can see, it is fixing the grammar and boom. Let's correct the grammar this time. And I think the app is working completely fine. By the way, if you want to refresh your app, you can also click here to refresh your app. So this is how you can build apps with Gemini three. 24. Examples of Google Gemini 3: Welcome back, everyone. So I'm going to show you some examples of Gemini three. If you go to ato do google.com, you can go to build. And if you scroll down, as you can see, guys, these are some examples. We have VaxllTybx. So I'm going to click here to access to this example. Let's break. Let's rebuild the eagle. And as you can see, guys, the app is working. Have a look. Now let's go back. Let's select this Shader Pilot. Let's click here to run this. And as you can see, guys, the app is running smoothly. So this is a showcase that you can make anything with Gemini three. Now we have this research visualization. Let's access to this. We can also view this info screen, and as you can see, guys, we have this website completely built with Gemini three. Now you can also change your device to mobile, and as you can see, guys, we have this smooth. We can change this to tablet, and we got this professional looking app. Now we have another example, bring anything to life. Gemini sees the code in everything. Drop in an old map, a floor plan, a diagram or a doodle, and what it turn into a working interactive experience instantly. So I have selected this image, and the gemini bring this image into life, and as you can see, guys, we can select anything. We have the name of that thing right here. So as you can see, guys, it brings the element into life. So you can select any image. This is the showcase of what gemini is capable of doing. Now if you scroll down, you can also access to more example. Let's browse the app gallery. And as you can see, guys, this is also generated by Gemini three P. I'm going to select this. Let's view this in full screen. And as you can see, guys, this is completely generated by Gemini three. You can also change your device as well, and this is a full interactive app. Let's go back. We also have this runner, which is also made completely in Gemini three initializing jump. And as you can see, guys, this is completely generated by Gemini three. Visual, decode, everything is generated by Gemini three. So the possibility is endless. The only limitation is your thoughts. You can imagine anything, and nowadays, you can create anything without having the knowledge of app making development, software, engineering, et cetera. Even a nuke can make apps and games with Gemini without having the knowledge of coding, et cetera. Now we also have this space shooter et cetera. We also have this square Metropolis, check this out. We also have this tempo strike which is also a three D game completely generated by Gemini three. We also have this image to Vauxll art. Let's view this. This is very interesting. You can also generate your image or you can select any of these example and you can zoom in, zoom out, rotate. W right click, you can control the camera. You can select this cat, zoom out, zoom in, control the camera, rotate the camera, et cetera. Let's zoom in, and I mean, check this out. Isn't this beautiful? We also have this kinetic shape, and we also have this SVJ generator. So these are some examples that people have created or built with Gemini three. 25. Free Access to Veo: Okay. Welcome back, everyone. If you want to access to VO for completely free without spending any money, first method is to go to asodio.google.com. You have to go to chat. Now go to Video and you can select 02. You can only access to 02 by using AIstudio and you can only generate ten videos. I believe it's a lifetime because this is only for testing. The second method is to go to flow.google.com. This is the website and you can click to Get Started. Now here, as you can see, I have already created project, but you can create a new project and you will get 100 credit per month. This is per month credit. You can generate five videos per month for completely free. Or you can generate one video if you want to generate quality video, which will take 100 credit per video. If you want to generate VO fast video, you can generate five video because it will take 20 credit per video. So this is a method to access to Veo for completely free. 26. Flow Overview | VEO 3.1: Welcome back, everyone. So this is the interface of law. If you want to create new project, you have to click on this new project. Once you create your project, you can rename your project by clicking onto this button Edit project. Now I'm going to name this to first project, press Center. And now as you can see, we rename the project. Now, if you have any videos, you can change the view. So let me go back. So if you click on this floor, you can go back to the homepage. You can open any project by clicking on it. You can change the view like this. As you can see, guys, now we have this view, now we have this big terminal view. You can also search for a clip. So I'm going to search for Wales presenter. As you can see, guys, it will only show us the video where we have the whales in the title in the prompt. Now, if you click here, it will show us all the favorite videos. Right now, I don't have any, so let's click here. Now let's set this video to favorite. So I'm going to click here. Now this is my favorite. And if I click here to the favorite, as you can see, guys, that video is going to show up. So let's go back. We favorite this video. Now we have to refresh the page. Now if we go back to the favorite, as you can see, guys, now this video will only show because we set this video to the favor. You can also click here to edit this video. You can also download this video by clicking onto this button. If you click here, you can flag your output or you can even delete it. You can do that by clicking onto this three dot. As you can see, we have flag output, we have Delete. Now, if you want to reuse your prom, you can click here and you can reuse your it will automatically copy and paste this fright. Now you can also preview this in a full screen like this. Now, if you want to find out how much credit you have, you can click onto your profile and the credit will be displayed right here. Right now, I don't have any credits. That is why we have this message. If you want to access to all the videos at one place, you can simply click onto your profile and you can go to library. As you can see, guys, now we have all the videos that we have ever generated. We also have the date of the videos. You can also access to the images. If you go to WIX, you can access to the WIX library. If you click here, you can go to the Image FX and you can access to the Image effects library. This is very helpful to find out your media. 27. Generate Videos Without Credits | VEO 3.1: Okay. Okay. Welcome back, everyone. Now if you have a pro account of Google One, you can generate videos for completely free without losing any credits. If I click here, as you can see, guys, I have 1,000 credits, which is for the Pcount, I can still generate videos without losing any credits. Let me show you how. First, go to gemini.google.com. Here, you have to click on this create Video. Now you can type prompt or you can upload your Image to generate image to video. Now I'm going to paste my prompt and I'm going to press Enter. As you can see, guys, my video is generate I have generated these videos, so this is my prompt, and this is my first video. Now, this is my second prompt, and this is my second video. Now for the third example, I use this image, which I have generated by using Image effcsO you can use Wix. I use this prompt, and We've got this lovely this time of year. You should see the dragons I met. I much prefer the classic tales myself. Now you can generate three video for completely free without losing any credits. If I go to flow, if I click here, as you can see guy, I still got 1,000 credit. Now the second method is to go to Google WIDS You have to go to Google Wits. You have to go to WIDS. You have to create new project and you can generate ten to 11 videos per day. And I'm not kidding you can generate ten to 11 videos per day. You can see, guys, this is six November date, and I have created several videos, and it did not give me any message that I have to the limit. As you can see, guys, I have generated so many videos. Whether you want to generate image studio or text studio, you can easily create by going to Video and as you can see, guys, it is using VO 3.1, but the resolution is 728 second with audio, not without the audio with audio. So you can generate ten to 11 videos with Google VDS as well. So if we combine these two, you can generate 14 to 13 videos per day without spending any credit. So this is the benefit of buying the P plan or flow that you can get 14 videos per day without losing any credits, and the quality is insane. If you look at the quality of any videos, the quality is insane. Check out. So this is how you can generate videos with the VO 3.1 without losing any credit. 28. How to Save Credits | VEO 3.1: Okay. Welcome back, everyone. Now I'm going to show you a trick to save credits so you don't have to spend your credits here and there you can follow this. If you have used image effects or whisk, you can generate age for completely free. Now, I have used this prompt, and we have this image. Now, if you look closely, the image and the video visual is pretty much the same. I've used the same prompt, and if I play my clip, as you can see guys, the visual of the image and the video is exactly the same. There is some minor differences, but as you can see, overall, the video and the picture is pretty much the same. So by following this trick, you can generate your image and you can visualize your video. We have another example. As you can see, guys, we have this prompt. We have this image. And if we go to the video, this is what we got. Now, you don't get 100% the same result, but you can get the idea how the prompt will look like. Now we have another example where I have used this prompt and I have generated this type of images. As you can see, guys, we got these kind of images. And if we go to the video, as you can see, guys, we got completely different video than the image. We have the same prompt. If you look closely, we have the same prom and we have the latest VO 3.1 model. If I play my clip, as you can see, guys, we have this type of video. So you don't get same result every time you use the image effects or is because these two are completely different model, View 3.1 and Image effects or image and four. Now we have another example. As you can see, guys, by using this prom, we got this image, and this is what we got by using text two with. Now we have another example where we have this person writing this creature. We have some other variation as well, and we got this prompt. And by using the same exact prompt, this is what we got from View 3.1. This time we have this prompt and we have this visual from Wisk or image and four. And in view 3.1, this is what we got. By using the same prompt, I did not change anything in my prompt. Now we have another version of the same prompt. In Gemini, I have this version. And I really like this version. So make sure you generate two version of your clip by using the Gemini. This way you won't lose any credit because you can generate three video per day for completely free. Now we got this kind of visual. This is my prompt. This is the bacteria pattern, as you can see, and with 3.1, this is what we got with the same exact prompt. Now, most of the videos visual are completely matching with the image. That is why I'm telling you to generate your image first, then generate your videos because that way you can visualize your videos. Sometime you will get this kind of result. So let me show you what I mean. So I have this prompt and this is our image. This is the second variation of this prompt. Now if we go to the floor, if I open this project by using the same exact prompt, if I play my clip, we got the visual, but as you can see, guys, we got just a Zoom in image effect. We don't get a full fledged video. So what I did, I just reuse this prompt in Google VDS and let me show you, and this is the video that we got. Now, as you can see, guys this time, we get this kind of video. So if you are getting this kind of result, then you can regenerate your video. That will help you to get your desired result. Now we have another time images example. As you can see, this is the prompt, this is the second variation. And if we go to the Vo 3.1 with the same exact prompt, this is what we got. Now we have this example where we have this prompt. This is my second variation. And if we go to the 3.1 result, this is what we got. Now we have this example. This is a neural network complex neural network. These are the images, and this is the video. So in most cases, the images are matching the visual of the video, but in some cases, the visual of the video is completely different than the images. Now we have this example. I'm going to show you the images of this prompt. So this is my images by using the same exact prompt, and this is what we got in full fledged video. I mean, check this out. You will get the idea how the video is going to turn by generating the images. We have the spiral, we have the spiral in both of these images. That is why you have to generate images first, then you have to generate your video. Now, here's the thing. If you are using the image and four, most of the complex prompt turned out to be like a concept art. So by using the same exact prompt, as you can see, guys, this looks like a real life example or game like video. Now we have this example with the same prompt. This is what we got from image and four. As you can see, this look like some cartoonish. But if we go to the Vo 3.1, this look like some kind of a scene. Now we got this image, and this is the prompt. We have the second variation of this prompt and with Vo 3.1, this is what we got. Now we have this example where we are hemming but this is the second variation, and if we look closely, this is what we got from a 3.1. I mean checkout the quality. The visual are completely matching with the image. Now we have this example. We have this prom. We have this visual. This is the second variation, and this is what we got from the 3.1. Now we have this example. This is the prompt, and this is the second variation. And if you look closely, the visual of the video is kind of matching the visual of the second variation. Now we have this prompt and this is the image. We have the second variation. And as I told you earlier, the image and four result are like a concept arch and the view 3.1 are like a real life example. The videos are photorealistic and the images are like concept art. This is the main difference between the Video and the imager. Now we have this example. We got this prom, and this is the images that we got from the same exact prompt, and this is what we got from View 3.1. Now this is the prompt, and this is the images, and we have the JCNPmpt of the simple English prompt. And this is the result of JCNPmp. And with the same exact prompt in simple English, this is the simple English result. As you can see, guys, the JCNPmp result is a little bit improved than simple English. Now we have this prompt, and I'm going to show you the simple English prompt. This is the prompt, and this is the JSON prompt of the same exact prompt. Now we have this example. This is my prompt. This is a JSMPm let me show you the simple English prom. This is the simple English prompt. We got these images and with the same prom, this is the video that we got from 3.1. Now we have this prompt, and this is the anime image. This is my second variation, and with the same exact prompt, this is what we got. Gunikansk. Now we got this prompt. This is the SN prompt. This is the second variation, and this is the simple English prompt. So I have used TSN prompt, and this is what we got. I mean, check out the visual. Now, if you are not getting the visual of Image and four from 3.1, then you can also use image to video. So I have used this image which I have generated by using Image and four. This is the prompt that I have used. And as you can see, guys, this is the video that we got. Now we have another example of image todo. As you can see, guys, this is my prom that I have used to generate this image. This is my second variation. And by using the same prompt, I have generated image TobloT is my image. This is the same prompt, and this is what we got. I mean, check out the visual. Isn't this beautiful? With the second variation with the same exact prom, this is what we got. So the view 3.1 did a pretty great job by animating these images. We have this image, and this is my prompt. As you can see, guys, you can pause the lecture and you can screenshot this, and this is what we got. Now this is my recommendation. If you are going for the image and four look, then you have to use the image and four image. You can turn that image into video because the view 3.1 is master at animating images. 29. Text-to-Video | VEO 3.1: Welcome back, everyone. Now I'm going to show you how we can generate text to video. So go to astudio.google.com and go to chat. Selah Gemini, and you can paste this prompt. Give me 50 ideas for generating images in Image and four. Let's click here. And as you can see, guys, we have plenty of idea. Now I'm going to copy this prompt. And first of all, let's go to Is and we can generate images. So that way, we can visualize our final video. You can see, guys, we have these images. Now let's go to flow. Let's create a new project, and now I'm going to select text video. Make sure you click here, you have to select your model. So I'm going to select Vo 3.1 because it will take 20 credits. And if you select VO 3.1 quality, it will take 100 credits. I'm going to select Vo 3.1 fast and I'm going to select one output. Can also change your expect ratio, but I'm going to select my landscape. Let's click here and we can paste this prompt. Now we have to simply press Inter or you can click on to this button. Now, while the VO 3.1 is generating the videos, we don't have to wait until the video is done. We can generate more videos. So now I'm going to copy my prompt. Let's paste this right here. Now we got this visual, so I'm going to copy my prompt. This is the second variation. Now let's paste this prompt right here and let's generate this video. Now we got these images, so I'm going to copy this prompt and let's paste this right here. All the videos are now generated. This is the first one. And now check out the visual. This is 3.1 first. This is the second example. Now, if you notice with the simple prompt, you can even get eight different scenes in one video. Now this is the final example and check out the result. So this is how you can generate text audio by using Vo 3.1. 30. How to Make Character Talk | VEO 3.1: Okay. Okay. Welcome back, everyone. So how can we make any character talk by using text to video by using Vo 3.1? First of all, we have to generate a prompt. So for that, we have to go to aistodio.google.com, clickhat. We have to select Gemini. Now I'm going to paste this prompt. So we have this basic idea. So this is the basic structure of your prom. If you want to generate any prom, this is the basic structure. If you want to make your character talk. Dinosaur with red Look doing a stand, we have this comma. Then I type his audience is also fellow dinos. We have Again, also write a simple joke that is easy to understand. After this comma, I have typed enhanced prompt. Now, after that, it just generate this image S type write prompt, and it just asked me to provide the topic. So I just copy this same prompt. And as you can see, guys, this is the prompt that we got, and this is the image that we got from this prompt. So I have used this same prompt, and this is what we got. Why don't dinosaurs ever play hide and seek? Because good luck hiding when you're 60 feet tall. Why don't dinosaurs ever play hide and seek? Because good luck hiding when you're 60 feet tall. I'm check of the result. Isn't this amazing? We even got the substitleO this dialogue box appear right here. Now we can also change the accent of your character. Now make another one in British accent. Now I'm going to copy this prompt and let's pase this right. So you can change your accent to any language, even British, Australian, American, et cetera. You can also change your accent as well. Now with the British accent, this is what we got. Right then. Why don't dinosaurs ever play hide and seek, eh? Because it's joy difficult to blend in when you're 60 feet tall in it. Right then. Now after this, what I did, I asked Gemini now Dino is playing Minecraft. He is getting angry. He can't complete it, then he start crying. Now, by the way, you have to add a comma if you are explaining something, if you are adding something new to your prompt. So that is why we have this coma because he is getting angry and after that, we have some different thing that I want my character to do. That is why we add a coma and a type, he can't complete it. Then we have another thing that is why we have these commas in my prompt. At, he start crying. Comma face came in a small box of the dino at the top right corner on top of gameplay. So this is the prompt that we got, and this is the reference image. So instead of going to flow, I save my credit, and I used Google VDS. And as you can see, guys, with the same exact prompt that we got from here, this is my prompt, and this is what we got. Are you kidding me? My house. Oh, no, not again. It looks like he is a YouTuber, even with the gaming chair. Now after that, I type also write some dialogues and write this prompt, and we have these dialogue. So what I did, I just copied the first dialogue and also I copied this prompt. This is the reference image that we got. With the dialogue, this is what we have. Why won't this cobblestone pillar just connect? Now, after this ITpes reacting face cam, a comma, he is happy, then suddenly he is in shock. This is the reference image, and this is our prompt. And this is the result that we got. Now instead of dinosaur, I create a new chat with Pixar Monster. Now I have typed this symbol prom. Pixar monster reacting face cam YouTube video. We have this comma. Then I type his expressions start from happy to sudden shocked. Add a comma, then I type right enhanced prompt. So this is the enhanced prompt, and this is the reference image t we got, and this is the result of 8.1 finally did it. Wait. What is happening? No way. Now you can also make two characters interact with each other by using text video. So I have type two Pixar monster joking each other, make simple joke, easy to understand, one liner roasting each other right enhance prompt. So this is the prompt that we got, and this is the reference image. Now, if we look closely in the image, we have a different joke than the video. So I'm going to show you a video. Then you will notice that we have a different joke in the video than the image. So in the image we have this jog, you are so slow, it takes you 1 hour to watch 10 minutes. The reason why we are getting different joke is because we did not write the joke. So the prompt says they are joking one liner, playful, one liner roast. So they are simply roasting each other with one liner. So that is why we have different jokes than the image. So this is what we got from video. They say the camera adds ten pounds, but I think you've just been eating the camera. Oh, yeah. Well, at least I don't use stilts to reach the snack machine, Professor Skinny. I mean, check out the result. The dialogue, the face, the lip syncing is amazing. So this is how you can make your character talks by using text Audio in w 3.1. 31. How to Create and Use Presets | VEO 3.1: Welcome back, everyone. So we have an interesting feature called Expand, and you can expand your prom to this preset. We have three pre made preset. One is cinematic, we have finire, we have action figure. Now you can also create your own customizable preset. First, let's test this preset, and then I'm going to show you how you can create your own preset. I'm going to type a simple prompt. So first of all, we have this simple prompt. A card is drifting. Now this is without any preset. Now I'm going to select action figure, preset. And as you can see, guys, in order to use this preset, we have to enter a prompt. So I'm going to use the same prompt, and then I'm going to select figure. And this time, the preset is going to expand my prompt according to the action figure. Now I'm going to click on this. Now let's use the original prompt. Now, make sure you use your original prompt. Don't use your preset prompt. Now I'm going to select film noi. Let's click here. Now I'm going to use my again, Spmpt and select cinematic. And now let's click on create. So as you can see, we got all the videos. So this is without any preset. Now, let's check out the action figure. I mean, check out the result. Isn't it look like the real life action figure? Now we have this film. Now we got some flaws in the videos, but as you can see, guys, the video look so much interesting. Now we have this cinematic Now, let's create our own preset. So we have to click on Expand and we have to click on. This Create New expander. Now, as you can see, guys, we have this instruction how to create our own customize or preset. So I'm going to create a two D animation preset, two D animator. Now I'm going to write my simple pro. Turn anything into two d Illustration animage. Now if you write your prompt, you have to click on this enhance so that it can meet this a preset. Now as you can see, guys, we have this preset. I'm going to click on this save so we can use this later. I'm going to write my prompt. A card is drifting. Let's select two d animator. And now let's run this command. I'm going to create another one stop motion. Now we have this simple prompt. We have to click on this enhance we have the preset, so I'm going to save this preset. Have our simple prompt. Now we can use our preset. Now, let's run it. So the videos are ready. Now check them out. This is with Tu Di animation preset. And I don't understand why it made this a three d style animation. I ask it to make a Tui, but we have this one. Now we have the store motion preset. And by looking at it study. The store motion preset work really well then the two D animation preset. So this is how you can expand your prompt and you can create your own customizable preset. 32. How to Generate Image-to-Video | VEO 3.1: Welcome back, everyone. So now I'm going to show you how we can generate image to video. But first of all, is 3.1 video is improved than the previous model. So we got this image, and I have this prompt. And as you can see, this is Vio result. Everything looks fantastic, but the camera moment is off the chart. But if I show you the result with the same prompt, with the same image, as you can see, guys, we got so much drama going on in this video. So 3.1 is improved in every single way then via symbol three. Now we have this new project. Let's switch from text to frame to video. Now we have two ways to generate frame to video. One is we can upload our own images. Second, we can generate images from scratch. So click here. You can upload your image from here. If you want to generate images, you can click here. Now we have this image and form model that we use in Gemini in Google AI Studio and in Wix as well. Now I'm going to paste this symbol prompt. Now if you click A, you can select your outputs. You can select your aspect ratio. You can also clear your prompt if you don't like it. Now let's generate these images. As you can see, we got this for result. So I really like this one. So we have this edited tools, so we can click here. We can type a text or we can press Enter, click here, do this, we can click here. We can change the color to any color. So I'm going to change my color to white. I'm going to select this stroke, and now I'm going to edit this image like this. We can select this image or we can select a particular area. Now, if you want to download this image, you can click here to download right now we are editing this image. If you want to create new image, you can click here and you can switch to create images, and you can create your images from scratch. But if you select an image, you can then edit that image. And as you can see, it is also using nano Banana. So I typed this prom, make selected area brighter, and outside selection, make that darker. Let's edit this image. Now, like the is in flow, we don't have any watermark, which is the best part. And now, as you can see, guys, we have this area brighter and we also had a selection is darker. Now we can edit this image in any way that we want. So I just make Galaxy more wider in purple. Let's donate. Now it did not change anything. So this time, I make Galaxy more bigger and glowing. Now if your prompt are not working quite well, so what we can do, in this case, we can download this image. Go to AI studio. You can select nano Banana, upload your image, and now we can make some changes. So I just make Galaxy covers the whole eyeball. Let's see if this time we can make these now have a look. This time we finally make this change. So now I'm going to download this image, go to flow. If you want to create new image, you can cancel this to create new image from scratch. Now I'm going to upload this image. Now, once you upload your image, we have to crop and select this according to the aspect ratio. Want to select this for portrait, you can select your portrait area. If you are selecting for landscape, then you have to crop and save this as a landscape. So let's select this area. And now we can type any prompt. So I just type the same prompt that we use to generate this image. Make sure we have selected 03.1 first. Otherwise, it will take 100 credits per video. And now we are generating image to video. One more thing, as you can see, guys, we have generated four to five images, but we did not get any image. The reason behind that is we did not download or we did not save any image. So let's let's go back to generate images. And let's paste the same prompt. Let's generate these images. Now, let's say you want to save any image that you want. So I really like this one so either we can download this image or we can save this as a new ingredient. If you click here, save this as a new ingredient. If we click here, as you can see, guys, the image is now uploading. And then it will save that image that we just saved as an ingredient to our library and check this out. We have this image in our library. Now let's check out the result. And I mean, the result is amazing. I mean, check out the detail Now I'm going to share with you some example that I have generated using 3.1 frame video. Now we have this example where we have this image, and I just type animate this image and have a look. Isn't this amazing? Now we got this car and I type card drifting, and by the way, this is the same prom data used to generate this image. I mean, check out the sound and the visual. Everything looks fantastic. Now, by the way, if you want to edit your image from the library, you can do that very easily. You have to simply click here, and let's say you want to edit this image using nano banana, you can simply let's edit this image. So I'm going to click here onto this pencil tool. You have to hover your mouse onto any image, click here. And now, as you can see, it is now using nano banana, and we can easily edit this image. So now I'm going to select this area. Let's select this area, and let's select I just remove all selected area and boom. As you can see, guys, it perfectly remove all the area that we have selected. So this way, you can easily edit your images inside this flow. Now we got this image, and I use this prompt to generate this image, and this is the result that we got from image to video using 3.1. Now we got this image, and by the way, I generated this image inside Wisk and this is the prompt data I used. And to generate this prompt, I used Google AI Studio. I asked Google AI Studio to give me 50 ideas to generate images in image and four. And one of the idea is this one. So I generate this image. And with the same prompt, we got this video. I mean, check out the visual. Isn't this look like a game card scene? Now we got this image, and by using the same prompt, I generate this image, and we got this interesting video. Isn't this look like a Subnautica knockoff? Now we got this and I edited this image using as you can see nanobanana inside AI studio. So let me quickly show you how I edited this image. So first, I start off with this image, Anatyp person holding logo with hand in the air, showing it close to the camera in the center. So with the same problem, I got multiple result. But this time, what I did, I make myself standing in a professional studio and cut to the final I use the logo, NSA person holding logo with hand, showing it close to the camera. With the multiple testing, this is the final result that I got. By the way, you can guide o to generate your videos according to your prompt. So I just type camera slowly zoom in to the logo and focus on logo. So I want o to focus only on my logo and check this out. It slowly zoom in and focus on the logo. Isn't this amazing now we got this image, Notype running in infinity hallway, and for some reason, the Google View can't accept this image. I don't know why. I got this error in Google Ds as well in flow and in this gemini. So what I did, I used this image note, meteor crashing on Earth, car running away from it. So I clearly type car running away from it. But for some reason, the car drives towards the meteor. Otherwise, everything looks fantastic. By the way, I generate this image using King. Then I use this pizza and type a hand slicing pizza into six pieces. To generate this image, you can simply type top view of a pizza. Then I got this video. I mean, check this out. Now, the slices are off the chart, but everything looks amazing. Now, I have generated some few examples so you can better understand how image to video work. First, we got this image of an apple and I type. If you click here, you can use your image. You can add your image from here and you can type any prompt into this dialog box. So I just type a hand holding knife cutting this apple into two pieces, and this is what we got. As you can see, guys, a hand holding knife cutting this apple into two pieces. So you can manipulate any image in any way that you want. Next, I got my image, this one, and I type person running very fast. And as you can see, guys, I'm running very fast. It changes the pace a little bit. I'm right now, I'm a completely different person than in this picture. As you can see, guys, these are totally different person, but it's close to my face. Next, with the same image, I type person running in a cyberpunk city wearing a cyberpunk suit. So I don't get any suit, but as you can see, I'm running in a cyberpunk city. So you can manipulate any image in any way that you want. Next, with the same image, I type person running away from a robot chasing him in the background, and this is what we got. Next, by using the same prompt, this is what we got. And in this example, I just type person running away from explosion happening in the background. Now in this picture, what I did, I try to test the expression. So I type person first nervous, start from nervous, sat, and then at the end, he smile and check this out. This is what we got. Now in this example, I type a prompt a slime falling from the sky covers the man body and check this out. Isn't this amazing now in this picture, what I did, I typed this simple prompt. The day is getting darker. Person set the head on fire, and the surrounding is lighting after. And this is what we got. It just made this completely new video. It did not just manipulate the image, but it just made this whole new video. So I did not like this at all. Now in this example, what I did with this image, if I show you right here, as you can see, with this image, I type a balloon covering the room, this type of prompt that I typed and check this out. This is what we got. Now in this image, what I did, I type day turning into night, add heavy raining, and check this out. We have so many flaws in this video. The video just made this whole new video without even manipulating my own image. Now in this example, I type volcano is exploding and the camera is heavily shaked as the volcano explode, and this is what we got. We have this anime style line shake. Now we have some more example. If you remember this image, we generated this image using Wisk and I typed the symbol prom. Both are fighting. And as you can see, guys, we got this incredible artisan. Isn't this amazing next, we got this burger picture, and I type a hand holding ketchup and adding ketchup on top of this burger, covering the burger. And as you can see, guys, this is what we got. So you can type anything that you want in your image. Now we got this example, and to make this image into video, I typed this symbol prom. The car is running very fast, and the tire is melting and check this out. This is what we call. We have melted tire. Next, we got this example to make this image I type a Ferrari car part at the edge of the mountain using Isc and for this video, I type a Ferrari car transforming into white and check this out. Next, we got this image N type something like this idea type, war fight or something like that, and this is what we got. Leave no survivors. Next, we got this Donut image, and I use which to generate this image. And with the same prompt, I generate this image. Let me show you. So for this car image, I typed this prom, cyberpunk style car. And for this image, as you can see, guys, we have red Ferrari car parked at the edge of the mountain. Now for this Donut image, we got this prom. And by using the same prom that we use to generate this image, I generate this video. As you can see, guys, it is animating this element beautiful, and we did not get any morphing next, we got this image, and I use this prompt to generate this image, and by using the same prompt, we got this video. Next, we got this prompt to generate this image and check this out. With the same prompt, I turn this into video. Now, this is what we got from this prompt. And with this same prompt, I turn this image into video. Have a look. Now at the end, I got this image, and this is a prompt that I used to generate this image. And I use the same prompt right here and I just add this additional line. He is trying to eat diver diver dodge his attack. So I type that this monster is trying to eat the diver, but the diver dodge his attack, and this is what we call. So this is how you can turn your images into video. 33. Generate Start and End Frame Videos | VEO 3.1: Okay. Okay. Welcome back, everyone. So now I'm going to show you how we can easily generate frame to frame video. So go to frame to videos. And if you notice we need two images to generate frame to frame video. If you upload one image, you can generate image to video, and in frame to frame video, we need two images. We need a start image, and we need an ending. This is where the video is going to start, and this is where it's going to end. Let me show you what I mean. So by using this prompt, I generate these multiple images. And in all of these images, I like this one. So what I did, I go to refine. And I type make this image darker, add cinematic lighting, also add studio lighting to the scene. And this is what we got. Now let's go to the define again, and I'm going to type gimme top angle view of this scene. Now, have a look. We have top angle view of this basket. I'm going to download this image. Now let's go to frame to frame. Now, this is the start point of our video, so I'm going to upload this one. Now we need the endpoint, so I'm going to upload this. Now you also need to type a prompt. So I'm going to use the same prompt that we used to generate. So I'm going to copy my prompt and I'm going to paste it right. Now let's switch to Video three fast, like landscapes, let one output, and click here. Now you can also swap the first frame with the last frame if you want to. If you did not like the first frame and you want the first frame, be your end frame, then you can click here to swap these two. And now the last frame is the first frame. And if you click again, now we have these two swapped again. So the video is ready. Let's view it and as you can see, guys, we got this smooth camera movement. Now, I don't know why it added this extra scene. I hope you get the idea how the frame to frame video work. Now, sometimes it also added additional scene like it added to this video. We have this example where I use this as my start image, and I use this as my last frame. This is the first frame, and this is the last frame. Check this out. So this was my last frame, as you can see, guys, so I don't know why it added these additional scenes. We have this example where we have this first frame, and at the last film, I am showing the logo. This is my image without showing the logo, and I use nano banana to generate this. And this is with the log. And now this is the final result. And by the way, I use this pro person showing logo on camera. Now we got this example. This is my first frame, and this is my last frame. And I type Apple Juice commercial, and this is what we got. So you can create multiple scenes by using frame to frame video, and it will cost you 20 credits. If you have to make a single frame video, then it will cost you 20 credit. But in frame to frame video, you can control your video. Now we have this example, and before I show you this example, let me show you how we can write prompt for the first frame and for the last frame videos. Now go to aistudio.google.com and you can type this prompt act as a start and end frame prompt, optimize it and write full fletch prompt. Now, as you can see, guys, it asks me to provide the prom so I type pizza fall from sky and lands on ground. This is my prom for the star frame, and this is my prom for the end frame. As you can see, guys, this is my first frame, and this is my last frame, and this is my symbol prom. These are far from the sky and lands on ground and have a look. So you can easily create start and end frame videos by using AI studio. You have to give this a simple prompt, and then it will give you the start and end frame prompt. Now, we got this example where I use my image, where I'm standing in this long grass, and we have this winter look as well. I type person waiting for signal, and this is what we. Now, as you can tell, we can generate multiple scenes. Next, we got this example where we have this car. So this is the image that I got from the whisk and I use nano Banana to remove the accident, and this is what we got. So I use this image, and this is my last frame, and I type car crash onto the next car, and this is what we got. So as you can tell, this way we can really control our videos. Now we have this example where we have these two fighting from a different camera angle, and this is what we got. Now we have this example, and if we go to is, if I scroll down, as you can see, guys, right here, you can really tell what is the prom because this is a really long prompt. But let me show you the simple prompt. Now to generate these type of interesting images, you can go to AI Studio. Now from here, you can type give me 50 ideas for generating images in image and four. You can get interesting prompt. So this is the prompt that I got from AI Studio, and as you can tell guys, at the end, I just make this closeup I use this prompt, and as you can see guy, this is my first frame, and this is my last. The way, you can also add this interesting line. If you want to make your start and end frame look better. If I go to my start and end frame, you can type this act as a start and end frame or prompt optimizer. Keep consistency and write full fledged prompt. You can also ask the Gemini to make a different camera angle so you can better make your start and end frame. Now for this video, I use the same prom that I used to generate this image. So as you can tell, we can really control our videos by using the start and end frame. Now in this example, I also use Google AI prompt to generate this image, but I also use nano Banana to remove the rocket. As you can see, guys, I type make rockets, stand still on ground. We have the rocket. Then I type remove rocket from the image, then a type add smoke after rocket launch and have a look. So I use this rocket lands on the ground as my first frame and the smoke image as my last frame. And I type rocket launching in space, and this is what we have. Check this out. So this way, you can really control your videos. Now at the end, I use nano banana to make both standing still face off. So as you can see, the robot is standing still, but the creature is not. So what I did make monster stand still. He stand sell, but mouth is still open. So I try make monster close his mouth, and this is the final image that we got. And to generate this monster image, I use the is. And by the way, this is also the AI studio prompt. I'm not that much intelligent to make this kind of prompt by myself. So this is a prompt that I got. And with the nanobnana result, I use this as my first frame, and this is the last frame. And I both are fighting, and this is what we have. Check this out. So this is how you can generate interesting frame to frame video. 34. Retro Image Editor | Google Gemini 3: Welcome back. One. Now go to astudo.cogle.com, go to build. You will get to this menu. I'm going to hide my navigation menu. Now I have typed this simple prompt. We are going to create a retro effect image editor. So I have typed this prom, build app retro effect for images. Let's build this app. So our app is ready. Let's upload an image. Now we have this image, and as you can see, we have basic adjustments like brightness. We have contrast, everything is working very well. We have the hue rotation. Now we have the retro effect. We have the CPA gray scale. We have this green green is working. Have this vignette. We also have this blur while everything is working. Now we also have this AI remix. If we go to AI remix, ats Blech. I mean, check this out. Isn't this amazing? We have this polaroid. Now we have this cyber punk. I mean, check out the result. We have this film, now we have this, I don't know what is called vapor wave, and finally we have this codec god. Now we can also click a start over, and it will reset your app. And finally, we have this codec gold. And now check. You can also add your own effect. You can apply your own effect, and you can reset the filter as well. Now let's start over. Now I also ask you to improve the preset and add more option to make retro effect. Now, this is the previous version, so you can click here to open your history. And as you can see, guys, improve preset and add more effect, this is the change this version. So I'm going to reset to this version. As you can see, guys, we have some additional effect, like we have this RGB shift. We have this pixelate effect. We have the scan line effect. If we go to AI remix, we also have some new effect. Like we have this one technicalar. We have this one, which is, I don't know what is type. We have this pixel art, and we have this last mo by the way, you can also add your own prompt if you want to. Now let's go to filter. Now, I also ask it to add classic 90s camcder effect. So let's go to the latest version. So this is my latest version. I'm going to select this and restore to this one. Now, as you can see, we have this camcoder effect, and we also have some quick look effect. So let's select 90s VHS. We have noir and we have film. We can also reset all filter, go to AIMx and we have these effects. By the way, you can select any effect. You can apply any effect to your images, and we have this download option. If I open my image, as you can see, guys, we have this high definition result. So this is how you can create a basic retro effect image editor. 35. 3D Audio Visualizer | Google Gemini 3: Welcome back, everyone. If you click here, you can hide your menu. If you click here, you can show your navigation menu. Now here, you can go to build, and as you can see, we are using Gemini three preview. We can hide this navigation menu. Now we are going to create a three D audio visualizer. So I'm going to type a basic prompt. Now as you can see, I have typed create audio visualizer of this audio in three D. Now we also have to add the file audio file. Now we have selected the audio file. Now I'm going to click A acknowledge. Now we have the audio file, and now let's click here to build this three D app. Now our app is ready. We also have to allow this. Now, as you can see, instead of creating the app with our audio, it gives us the option to use the microphone. So I'm going to click here to use the microphone. Hey, Whats up, guys? My name is Faizan Amjed. And now, as you can see, guys, we have this three D audio visualizer. We can move this around. We also have an option to upload and audio, so I'm going to click here. Welcome back, everyone. So now we can access to Nano Banana Pro if you have the Pro account of Google. You have to simply click here and as can stop the audio. We have a new option. We can play the audio. We can zoom in and we can zoom. Now we can also refine this app even more. So I'm going to type a simple prompt. Polish design, addEma and then I type at cinematic camera control. Now we have this updated version. Now I'm going to select my microphone. Hey, Whats up guys. And as you can see, guys, this is the updated app. We have the polish interface. Now I'm going to stop the recording. We also have an option to toga cinematic camera, so I'm going to click as you can see now we can move this around. Let's click here and check. Isn't this amazing? Let's stop this and now I'm going to make some changes. Now I have type add more design for three D audio visualizer, so I want more design, more variety of this three D audio visualizer. So now let's send this prompt. Now, for some reason, if we enable the microphone, we can't really see the three D audio visualizer. But I'm going to upload my audio file Welcome back, everyone. Now, as you can see, guys, we have this three D audio visualizer, and we have four design. As you can see, I ask you to add four design. Now we have four design. This is design number one. So now we can access to Nano Banana. This is design number two if you of Google. This is design number three. You have to simply click here and A and this is design number four. Option called. Now if we switch from file upload to Microphone, as you can see, now we can as you can see, now we have this three D audio visualizer. We can turn the camera off. We can switch from different design and now have a look. Now let's ask to fix this problem, so I'm going to ask it to polish the design. So I ask it to polish the design and make it unreal five rando design. Now, if you face any error, you will get this message at to fix. We can click here to fix the error. After we fix the error, as you can see, we have the policy design, and if I click on the microphone, still we got the problem. Now let's upload the file. Welcome back, everyone. So now I mean, check out the design. Isn't this amazing to Nano Banana Pro if you have the pro. Let's switch to grind. Count of Google. You have to simply click here, and as you can see, we have a new option called Create Image if you select. Let's select the wave. This we have Nano Banana Pro. Now, on the other hand, if you have the free account, you will access to the previous version of Nano Banana. Now, let's switch to core. But if you have the pro version, you can access to Nano Banana Pro. Now, to use this Nano I mean, check out the design. Isn't this amazing? Let's switch to microphone and have a look. We have this beautiful design. We can control this with the mouse. We can zoom in, zoom out. Let's zoom in all the way. Now we are inside or I mean, check out. Isn't this amazing? Now I will give you a short assignment. I want you to fix the problem if you having the blank screen. If you click on Microphone, as you can see, guys, we don't have any design, but if you upload the audio, you will get the designs for three D audio visualizer. I want you to ask Gemini to fix this microphone problem. If you are facing this problem, I want you to fix this problem by typing a simple prompt. Screen went blank when using microphone, fix this problem. Just type the simple prompt and see if you can fix this problem or not. This is for your assignment. 36. Profile Audio Visualizer | Google Gemini 3: Welcome back, everyone. Now we are going to create a simple app called Profile audio visualizer. In previous lecture, we have created three D visualizer for audio. This time, we are going to create profile visualizer for audio. Now I have typed this simple prom. Build an app input profile and make it circle audio visualizer around profile in a plain background. Let's build this simple app. Now here's the thing. If you want to build the perfect app, start from the basic, then you will improve your app step by step. Don't type a large prompt or don't type a paragraph prompt to make app in one single prompt. Just build your app step by step, build the basic structure for your app, then you can polish your app afterward. Now the app is ready. We need to allow for microphone. Now as you can see, we can choose the profile. So I'm going to click here to change my profile. Now we have the profile, but we can't really crob or we can select the profile area. I'm going to click here. Hey, what's up guys. Hey, What's up guys. My name is Faizan Amjed Hello, hello, one, 23, mic testing. As you can see, guys, the audio visualizer is not working. So I'm going to click here to cancel my microphone. So I've typed this simple prompt. Visualizer for audio is not working. Fix that. Let's send this prompt. Now the array is fixed, and as you can see guys, the issue was that the audio analyzed node created inside this int and stored in a react p was not triggering. So that is the reason why the audio visualizer was not working. Et's add the profile and let's unable what's up guys? Hello, one, two, three, M testing. And still we are not getting the audio visualizer. So I'm going to click here. So now I have to have this symbol prom, add option to adjust profile, add an option for choosing color of visualizer and background. Add input audio option as well. Let's send this command, and hopefully after this prompt, we will get our audio visualizer. Sometimes if you are getting error and you are still getting the same error, try to build a new project with the same prompt. As you can see, guys, with the same prompt, I did not tweak any words here and there. And as you can see, guys, if I select this star live chat, the audio visualizer is now working. Now I'm going to click here to choose my profile. And as you can see, guys, we have the profile. Got it. Now let's improve this app. Now we have this prompt add option to adjust profile, add an option for choosing color of visualizer and background. Let's send this prompt. Now we have the updated version of this app, so I'm going to select my profile. I'm going to click here. And as you can see, this is the visualizer theme. I'm going to click here to start live chat. Hey, What's, guys? I'm going to change my theme and as you can see, guys, all the things are working perfectly. We can change the background to plain color like this. We can adjust the profile size and now, as you can see, guys, right now, we have this option to start live chat. So instead of the live chat, I ask you to make audio visualizer. Now we have the updated version. Instead of the live chat, now we have the simple audio visualizer. Now we can symbolically care, we can change the theme color. We can also upload the profile. Here we have the profile. We can adjust the profile size. Everything is working. I ask you to add four type of visualizer. Now we have four designs. So first, I'm going to change my profile. I'm going to select this profile. Let's start audio visualizer. The audio visualizer is working. We have the bars. We have the waves, we have arcs, and we have pulse. I mean, check this out. We also have profile size. We can change the theme. Everything is working perfectly fine. I really like this one, which is by default, the bars. Now I also ask you to add cinematic camera controls. So the Gemini is currently working on it right now. Now we have the cinematic camera control. Now if I click here, as you can see, guys, we have the orb. Hey, what's up, guys? My name is Faizan md. This is the static. I'm going to increase the size of my profile. This is the orbit. And as you can see, guys, we have this smooth cinematic camera. We have the gyro, and I'm in check this. The profile is moving with my cursor. If I select the orbit, I'm in check this out. He just making this orbit animation. And if I select the gyro, we have this profile moment when I'm moving my mouse. Here's the best part. You can minimize your control. You can change this to full screen and have a look. You can make this YouTube video. You can change your profile color from here. And as you can see, if I select this profile, as you can see, guys, you can make this a YouTube video. You don't have to animate anything. This is live recording. You can record this live. So this is how you can make profile audio visualizer using Gemini three P. 37. Caption Generator App For Videos | Google Gemini 3: Welcome back, everyone. Now go to astudo.google.com. Go to build. You can click here to hide your navigation menu. Now, we are going to build a caption app. So I have typed this symbol app, build an app that input audio and make caption in greens. Now we can click here to build this app. Now, as you can see, we have this app. We have this menu where we can change the color of the background. We have the background color. We have the font. We have the phone size. We have the phone weight which is normal and bold. We also have text color, and we also have shadow, and we got alignment as well. Now as you can see right now, it can only access to my microphone and it can generate captions from it. And as you can see, all the captions are now written on the screen in green screen. We can change the color of the caption. We can also disable and enable the shadow, as well. Now, I don't know why it just type hind Now I'm going to click AT stop my as you can see, we don't have any option to upload the audio. So now we can type at option to upload audio, and as you can see, I have already done that. That mean just go to my previous version. You can also go to your previous version of your app. From here, you can open version history. Now, from here, you can select your version. Now I'm going to restore to this version. And now as you can see, we have an option to upload the audio. Now I have selected this audio. I'm going to play this audio. Welcome back, everyone. So now we can access. And as you can see, we have this caption on the screen. Simple click here, and as you can see, we have a new option called Create Image. Now I'm going to stop. Welcome back, A. Now, next, what I did, I asked you to add a download button for video to export. So I want my caption to export in a green screen. Now, as you can see, it added this sport button, Export Video, so I'm going to click here. And if you notice it ask my permission to record the screen, so I'm going to allow this. And as you can see, guys, the Gemini is now recording my screen so that after in production, I can remove the green screen and I can add the caption. Now we can stop sharing from here. Now I'm going to click here on the option. I'm going to stop this audio. Now, next, what I did, I asked Gemini, add animated caption, add four animated caption option to choose from. And now, if you notice, we have the animated style. So I'm going to upload my audio. I'm going to play my audio welcome back everyone. So now we can access. And as you can see, this is the pop animated caption. We have slide. As you can see, we have a new option called Create Image. We have bound. We have Nano Banana. We have fat on the other hand. And we have none. Now, as you can see, all of these caption are coming at once. So next, I ask Gemini add option to make captions appear word by word. And now, as you can see, we have the option. We can click here for full sentence caption, and we can click here to add caption word by word. So I'm going to select my audio play. Welcome back, everyone. So now access to. And as you can see, this is the full sentence. You have Word. If we click Create image. Check this out. Like this, we have Nano Banana Pro. Now, on the other hand, if you have the Let's stop. Now, in final, I ask Gemini to add option to edit transcript. So if the Gemini get my word wrong from my audio, we can fix that afterward in transcript. So now I'm going to upload my audio. I'm going to play only. Welcome back, everyone. So now we can access to Nano Now if I click to stop my audio, we can close the option and as you can see Come back everyone. So now we can access to Nano Banana. Drop this, click here. We can double click and we can chain caption on the previous. Isn't this amazing? So this is how you can build a basic app for captions using Gemini three pro. 38. Lipsync App For Images | Google Gemini 3: Welcome back, everyone. Now, go to asodio.cool.com, go to build and you can click A to hide your Navigation menu. Now we are using Gemini three P. I have typed this simple prom. We are going to create a lipsync app for character. So I type this prom, build a lipsync app to photo for animation, ask for photo and then audio, then lipsync wrong. So we are going to animate photo by using the audio. Now we can click here to build this app. Now we have this app, and we need to select our character for lipsync animation. Now I have this pixel art character. I'm going to select my mouth. So right here we have some instructions like click and draw to draw a box around the closed mouth, so we have this mouth, draw this like. Now confirm the mouth. Now we need an audio. Now, the gemini is listening and analyzing the audio, and now we have this Welcome Animation. I mean checkout the result. Google, you have to click. We have a lipsync character animation. Now, next I add this feature, add option to make I motion. So as you can see, the character eyes is just neutral. They don't do anything, so I just add it I motion. And if we go to the previous version, now we need to select the eye. I'm going to select my left eye, next eye, and now I'm going to select my right. Now if I play my animation to master Nano Brenna pro Check this out. I animation as well. Everything. Next, I say, add a download button to download lipsync video. Now we need to select the mouth, we need to select left eye, and we need to select the right. Now, as you can see, we have this download video. So if you click on it. First, it will record your audio, then it will download your audio. Is for you. Now check this out. We have this download video. Hi, if you want to master Nano Brenna Pro. Now next I add option for emotion control and now have a look. In the latest version, we have this animation control. So if I play my invasion, this is neutral. This is happy. This is angry. And as you can see, the mouth movement is aggressively increased. You have surprise. Now this is your mini assignment. You need to add an option to change your audio because right now we don't have any option for audio replacement. If we click here, it will again ask us for an avatar or for a portrait, and we need to select the mouth, eyes, and audio again. So this is for your mini assignment. Add an option to this app for audio replacement. Also add an option for character replacement as well, and you need to add another option. You need a zoom in and zoom out, which I also added right here. But as you can see, this feature is not reliable. If I zoom in, as you can see, guys, it is zooming in to this position. So you need to add a feature to zoom in and drag your screen to select your mouth. So you need to add these three features. This is your mini assignment for this app project. And this is how you can create a basic lipsync imeter using Gemini three pro. 39. Map Location Generator | Google Gemini 3: Come back, everyone. So go to astodo.google.com. You can go to build. You can click A to hide your navigation menu. Now, we are going to build a map simulation. So I have typed this simple app, build a app simulator, ask for city name at cameras smooth motion, at particle, and show space view of Earth. Now we can click here to build this app. Now, as you can see, the app is ready. We can type any city that we want. So I'm going to type New York, I'm going to press Int and as you can see it highlight the New York City on the map. I want the gemini to add camera control and highlight the city that I have searched for. So I just ask add camera Control, highlight city. Now we got this update, so I'm going to search my city name. I'm going to press Inter and check this out. We have this highlight city effect going on. Now next, I want a handheld camera motion, and I want a polished design. So I just ask add handheld camera motion and polish design. So now we can click here to make changes. And now, as you can see, we have this handheld camera motion, and we have this polish design. So let's search for the city name, press Center, and have a look. Now, as you can see, guys, the camera is moving way too fast. We don't have any smooth camera motion. Now, we ask for polished design and a handheld camera motion. So we got this change. Now I'm going to search for a city. I'm going to press. Now, currently, we have a problem. We don't have any city name that is located on this map. So next I ask highlight located city. Now we got this version, so let's search for the city. Let's press Enter. And now, as you can see, we have the city name. You can search for any city that you want. Let's press Inter. We have the Mexico and check this. We have this highlight next, I ask Gemini at camera movement like Zoom in and Zoom out pan, et cetera, and we got this update. As you can see, we have these camera controls. Let's search for this city and now have a look. We have this zoom in. We have this Zoom out. We can control the camera, we can use the scroll to zoom in and zoom out. Now if you try to generate the same app in a new project using Gemini three Pro, you will get a different result. As you can see, guys, I have the exact same prompt that we use to generate the map simulator. And if I type my city, as you can see, we will get different result and check this out. We have different results. If we go to the open version history, let's click here, and this is my latest one. I'm going to restore to this version and now check this out. We have the camera controls, and now if I type my City press Enter, check this out. We have this New York City. This is Zoom Out. If you click here, this is Zoom Out. This is Zoom in. And if you click here, this is camera Controls. If you click and hold it, you will get this animation and for this. So you will get different result if you created a brand new project with the exact same. 40. Productivity Apps | Google Gemini 3: Welcome back, everyone. So now I'm going to show you how we can build productivity apps. So I have three examples. First one with a single prompt, I have to build an app, 25 minute timer clock alarm that alert in custom time. Now, as you can see, we have this timer. So if we start, we have this countdown. So this is basically, I believe a Chinese trick that a 25 minute clock, and you can focus on your work in 25 minutes. So you don't have to do anything in between 25 minutes. Now we can pause this time. We can go to short break, five minute short break, we can go to L or if you want a custom duration, which I ask for custom time, we can click here and we can type custom timing. Select the time, click Okay, go to this. We have to select the custom time here. Now, click and now have a look. We have 20 minutes running. Let's close this, and this is the second app. With a single prompt, build an app to do list, we have this app. Now I'm going to type drink glass of water. This is task number one presenter ply second, go to work. Now if we click here, as you can see, guys, we have this line, it means the task is done. If we go to complete it, this is the completed task. If we go to active, we don't have any task. So let's go to all and if I uncheck this, go to active, this is the active task that we have to do. Now we also have these options. You can either delete. You can also break down into subtask with AI. So let's click here. And as you can see, guys, we have this short task of this long task. Now we can also enhance this clarity of this task. So if you click here, check this out. We have this start. So with a single prom with some two word, you can build a complete app for yourself. Now, this is the final productivity app. Build an app time management app that take task as text, and you can set timing for that task. Now, have a look. We have the task, go to work. We also have the timing. We also have a SMAT plan. So let's click here add and we can also click here onto Smart Plan. Now we have to enter go to work with the same text. I'm going to click here Smart Plan. Now, have a look, guys. With the Smart plan, we have these mini task of this main task. Now, if you click on it onto any task, we have the timer for that task. So let's start timer and check this out. Wake up and complete morning routine, get dressed and have breakfast, et cetera. We can click here to go to work. This means we completed a task. If you go to task, we have this line up here. We can delete this task. If we click here, we have the timer for this task. If you go to stat, check this out. We have one pending, which is this one. So this is how you can build basic productivity app with Gemini three P. 41. Flat Vector Avatar App | Google Gemini 3: Welcome back, everyone. Now we are going to create a basic app for our third generator. So go to ato.google.com, go to build and you can click here to hide your navigation menu. Now we have this basic prom, build an app for two d vector Cc creation. Take inspiration from this reference. So we need to add a I'm going to add my reference. You can find to the vector arrector from Google Images. Now we can click here to build this app. Now, as you can see, guys, our app is ready. We can select the skin tone from here. We can select the shape of the face from here. We can select the nose. I really like this. And also randomize your character. We have the hair. We have color for hair. We have mustache, we have beard, et cetera. We also have features like wink, white, happy, we have laugh. We have glasses like round. We also have some cloth and we also have the background. Now you can download this SIG. Now, next, I want my character to turn 360 degree. So I type this pro Make character turn 360 degree using slider, so I'm going to click here to open my previous version. Now instead of character 360 degree moment, it just move the whole. Can fix this problem. So this is for your task. You have to fix this problem. Make the character move 360 degree, not the whole card. So just type your prom to fix this problem. This is your mini assignment. Next, I type this simple prompt, fix the position of the hair, also add hands and foot and add hand and feet motion control. And now if we go to the version history, I'm going to my next version. Now let's go to the pose, and as you can see, we have this arm pose. We have this left arm, and we have this walking. We also have this wave now next I type redesign and Polish app add shading. And if we go to the app, I'm going to select my version from here, restore to version, and check this out. We have this new design. Next, I type, add a full body pulse control using slider like thinking, angry, et cetera. And if we go to the final version, which is this one, restore, let's go to the pause, and as you can see, we have angry, confidenc thinking and we have neutral. And final, I type fix here at close up and add camera control. Now this is my latest app, so I'm going to select this and I'm going to restore and I add a close up camera control, and if we go to the view, as you can see, we can zoom in. We can use pan vertical pan horizontal or we can click here to add a close up. And now, as you can see, we can change the color of the skin tone, change the color of the face shape. We can also add this nose. We can also add this here. Check this. So this is how you can create a basic vector Avatar app using Gemini three Pro. 42. Building Games Clones | Google Gemini 3: Welcome back, everyone. Now I'm going to show you how we can build Clone Games app using Gemini three P. So I have two examples. Now, first, go to Google AI studio, go to Build and click here on to Start. You can click here to hide your navigation menu. Now I have this simple prom build a game for limbo style platformer. We can click here to build this app. I have already done that and check this out. We have this basic limbo game app. Now, as you can see, we can move around, we can jump. This is a platform. Now, next what I say, add anime, addFreflies, particle basically, you have to type your changes right here and you can click here to send prompt to make changes to your game. Now here we have the enemy fireflies and particle version. As you can see, we have these spiders. We can avoid these spiders as enemy. If we touch any enemy or obstacle, the game is end. Next, what I say, make jump a little longer. Add a gun, make character a person, and now have a look. We have this person, we have the gun, and as you can see, these are game control. These are for movement. This is for jump, and this is for the shoot. So if I press X, as you can see, my character shoot, now we have a longer jump as well. So we can do plot Former very easily. And as you can see, this is 100% working. Next I say define character animation, add a crouch option, add a boss fight, and now have a look. We have the animation for character, we can crouch. But as you can see, guys, we have this error. My character, when I crouch, it does fall down. So next I say Wix error, when crouch character, fall, and now have a look. We don't have that error anymore. We can crouch very easily. We can jump, move around with X, we can shoot, now have a look. We can shoot enemy, and we also have the boss. And once you finish the boss, the game end. So this is a basic clone for limbo. I'm going to show you an example of Mario. Now for Mario game, I use gemini.google.com, and I have the pro account of Gemini. So I use this pn, built Mario style game and have a look. We have a basic Mario style game. We can move. We have this basic game. Next, I fixed platform can't jump high. So as you can see, I can't really jump high, so I ask for this to fix. And in the next version, let's start the game. As you can see, now we can jump a little longer. So if you are facing any problem, you can ask Gemini to fix that problem, and the gemini will fix that problem very easily. Now, next, I say, add more speed to the character. So currently, as you can see, the speed is very slow. So I want my character to move very fast. So in the next version, check this out. We have the increased speed version. And have a look. We have this dying character animation as well. Next, I say, add cinematic light and shadow, and if I start my game, have. Now, in my next version, I ask for redesign and polish game and have a look. If I play this game, we don't have anything. So I just type screen went black fix it, and in my next version, Gemini fix this problem. And still, I got this error. So I click here Fix error, and Gemini fix that error as well and have a look. Now we have this game. This is the Polish version of our Mario style prototype. Mm and now have a look. If you complete your game, we have this score as well. Now, final, I say, make cinematic cameras smooth motion, also increase speed. Make level harder level in the world. So in my final version, this is my finished game. As you can see, guys, this is a little harder than the previous version. Now, as you can see, we can't really have this coin animation. So you can also ask for this to fix, but I did not ask for that. So this is how we can create basic loans for games using Gemini. 43. Build Mini Games | Google Gemini 3: Welcome back, everyone. So now I'm going to show you how we can build complete game from scratch. So go to astudo do google.com. If you go to playground, you can type any prompt. You can use Gemini three pro for prompt. Now I have typed this simple prompt. Give me Retro 60 bit five puzzle games idea for Gemini three P, give prompt. And as you can see, the Gemini give me these details prompt. Now you can also come up with your own idea. You can see, guys, I have typed puzzle game, hidden cat cat hiding in a room full of stuff. This is my final version. So I'm going to show you the first version. So this is my original idea. Let's restore to this version. And as you can see, with the basic prompt, we got this puzzle game. And now have a look. We have to find the cat. We can zoom out. We can scroll up. We can drag this to this side, and we can find the also have the timer. If you can't really find the cat, the game is over. Now here we have the cat, so I'm going to click here. This is my cat. And as you can see, the game is not working very well. So in my next example, I type make it three D game in a room. Let's go to previous version, and I'm going to switch to this version. Let's select space station, and now have a look. We have this three D room where we can find. So this is very difficult to find the cat in this version. Now next I typed now make it two deep pixel 60 bit game style. And this is my final version, so I'm going to click here. This is my final version, so I'm going to select that, restore to this version, and now have a look. We have to select the room. We have to select the zone, so I'm going to select magic forest. And as you can see, in 60 bit, the games work very well and looks very good. And now check this out. Now, we also have the hint where the cat is hiding and check this out. We find that level is clear, we can go to next stage. This time, I'm going to select dungeon and hea the cat. Accidentally find it. Now, this is my original idea, but as you can see, guys, the Gemini give us these prompts, so I use this prompt. I copy from the concept to the code prompt copy. You can go to build. So let's go back. Let's go to build, and we have the start. We have this stem 93. You can paste your prompt right here. And you can build your app. So I have already done that. So I'm going to show you this example that the gemini give me, and I'm going to show you the final results of these prompts. Now we have this example, Farm simulation and the gemini give me this prompt. So I basically paste the Gemini prompt, and this is one single prompt. And as you can see, we have this complete game. We can click here to launch the game. We have the defense mechanism, and we can use them to place them here and there, and we can ready for the defense is basically a clone of plant versus zombie. As you can see, guys, now we can reboot, we can refresh the app, and we can deploy this. We also have the credit left. So this is basically how much credit we have left. Now we can value the fence. Now, this is basically one single prompt. If you want to take this to another level, you can ask for the polished version, redesign version, or you can ask for improvement in this game. Now, if you use Gemini or building your game, you might get some error. So I have used the Gemini prompt, and as you can see, we have this space shooter game. I can move around with my arrow key with the space key, I can shoot the enemy. And if I touch any enemy, the game is end. So this is basically a game. Now I just improve a game a little bit. If I scroll down, as you can see, I have typed, you can type your changes from here and you can send your prompt from here. Polish game and make it 90s Arcade class. Now, if you click here, you can go to your previous history. If you click here, as you can see, I have make five changes to this game. So this was my first change. Now I'm going to select my second change. Now let's go to restore to this version and check this out. With the 90s Arcade style, the game looks so much interesting and is fun to play. Check this out. We have the now, right now in the game, we have only these mini anim. We can shoot these enemies, and if I touch any enemy, the game end. So next I add a boss wide by simply typing, add a booze. If I click here, go to my previous history. If I scroll down, this is my third change, so I'm going to select that, and I'm going to restore to this change. If I play my game, I can show these enemies. We have a warning that a massive energy is detected, and the boss fight is also a little difficult to beat. Now, next, I type, add plane selection, add more weapons, and add five more level with separate. Each is different and powerful than each other. So I'm going to select that version, which is this one, and I'm going to restore to this one. Now, by the way, if you notice I have opened these changes, the previous version. So we have all the list right here. If you open your previous version a while ago, your days and the timing will appear right. So be careful when you selected that. Make sure you selected your previous version from at the bottom of this drop down menu. Now we can restore this version. And if I play my game, we have the selection screen. We can select any plane. We can launch the game and have a look. This is my first version, and we touch maybe these enemies. That is why we have this error, so I'm going to insert the coin. If you find error, you can simply refresh your game. You can start. You can select your plane, launch the game, and we can avoid these enemies, shoot these enemies as well. By the way, this is level one, and we have the boss fight boss fight. And once you touch any of this bullet the game now final I say game keep crashing fixed. So my game was crashing, so that is why I ask for changes. If your game is crashing midway or if you find any error, then you can type your changes. You can ask for fix that problem and the gemini will fix that problem. Now we have this example. With the Gemini prompt, we have this game. We have to rotate and we have to make this person find this. So you can rotate and you can land this person inside the spoke. Now maybe it look very easy, but it is not. As you can see, it's very difficult to find this. I have this example. As you can see, guys, with the Gemini prompt, I did not type any prompt at all. With the Gemini prompt, we have if I initialize this sequence. Basically, the concept of this game is that in 10 seconds, you will have a clone. So this is a chronoclon game. Now I'm going to type initiate sequence. We can go back. We can generate any sequence. So that is a fun part. You can generate multiple sequence, and by the way, we have the limit max clones three loop is 10 seconds. So after 10 seconds, you will have your nucleon you will get three clones. Now we have this sequence, and as you can see, guys, after 10 seconds, the game is reset, we have the loop, and this person is doing what I did last time and now have a look. So this is a very fun game. You can take this game to a further level if you want to. Now, next we have this game. We have to mix these liquids, and this is also the Gemini prompt. I did not type any word at all by myself. Is basically the tetrus for the liquid. Now we have this next so we can drop. We can mix these liquids. So this is blue, so I'm going to land this right here. This is the red one, so I believe goes this way. Now we have the yellow and the red. So I'm going to land the yellow. We have the red, so I'm going to drop this on. So this is a basic. Finally, we have this game with the Gemini prompt. As you can see, I did not type anything except the game was not working, so I type game is not working, fix it. And the gemini fix that problem, and my game is now working. Now if I re try injection, as you can see, the game is very simple. We have to connect this. We have to align these, and we have to make puzzle. So you get the idea. So basically, this is how we can build mini games using gemini from scratch to finish. 44. Build Boxing Game | Google Gemini 3: Welcome back, everyone. So using the Gemini, I am building a boxing game and using the AI studio, I'm building the same game. So I typed this simple prom, build game boxing game, 60 bit pixel, and this is what we got. We can press Z to start and check this out. We can block with X, and we can punch this person with Z. But as you can see guy, this person is not fighting back, so we have this problem. So X for make it modern remastered. And if I go to my next version, check we have this Zw box. We can press X to block any punch. But as you can see, guys, this person is also not fighting back again. So we got this error. But I click A to fix this error. And in my final version, as you can see, guys, now we have this fully functional game. We can punch this person. We also have this plot animation. We have this block animation. We can knock out this person as well. We can move around. Let's punch this person. Let's get out now you get the idea. We build this using just two prompts. Now I build this exact same game using the Google AI studio. So I typed this prom build game boxing game, 60 bit pixel character can move around in boxing ring, add action, blocking, punching, uppercut, et cetera. Now, as you can see guy, this is a basic game. We can move around, we can move forward and backward. And if you win, you got this message. Now, next, I ask for make it modern remaster. And have a look. We have modern remaster. We have this key. If you want to move around, you can use this key for quick jab, you can use JF block, you can use S for heavy uppercut, you can use K. And now have a look. We have this design we can block, we can punch. You can also use that key, so I'm using my Z key to punch this person. Now next I ask Demni polish and redesign game, add moving sideways and upward, downward character movement. So I want my character to move around in the whole ring, not just sideways. So I ask Demni character movement. Now our character can move around in the whole ring. We can punch this person. Now in the final, I ask D and I make it 60 bit pixel style redesign characters. Now this is my final version of boxing game. And as you can see in the background, we have crowd animation. We can move around in the whole ring. We can beat this person. And if you notice, guys, if I go back to this ring, my character design is getting smaller because I'm getting away from the camera. And if I come close to the camera, my character design is getting a little bigger because we are coming close to the camera. Now let's beat this person. Now you can also polish this design as well if you want to, but I really like this design. You can make your entire game just by using one single prompt as well. So this is just up to your imagination. So you can build any game, any app, any design that you want. Using Gemini. 45. Build GTA 1 Game Clone | Google Gemini 3: Welcome back everyone. Now I'm going to show you how we can make a fully functional full fledged, finish the game. So I typed this basic prompt. Build a GT one game clone, just a dam of fully functional. Now, I have used geminigole.com, and by using Gemini three pro I have created this game. And as you can see, guys, this is a fully functional game. We can move around. We can drive any car. We can get out from this car. We can steal another car. We also have these keyboard keys to do anything in the game. Now with Google atudio.google.com, with the exact same prompt, I have a different output. Check this out. Have the character move around. We also have the cars. So let's go this way to find any car. Here we have this car. And the design is really bizarre because it looked like two cars are merged together to create a new one. Now, next, I ask Gemini to polish my game, make it more functional. Also design a simple 2.5 d City. Now I'm going to go to my next version. As you can see, guys, we got this error. I click here to fix this error and after that, I'm going to show you my next version and have a look. We also have the animation for character. Now we can press F to steal any car, move around in the city. We can get out from any car. We can explore the city. Now I do the exact same thing to the AI studio. Polish game, make it more functional. Also design a simple 2.5 d city. Let's go to the next version. Now have a look. We have this game. But for some reason, my character is stuck in somewhere. But as you can see, guys, we have this design for the game. You can ask Gemini to fix any error if you are facing any. Now, next version, what I did, I asked Gemni to make it to the redesign game, add controllable Zoom. Now, I want zoom in and Zoom out feature in my game, so that is why I ask Gemini for controllable Zoom. Let's go to the next version. And as you can see, we have zoom in and we have zoom out using the mouse scroll. Now I can move around in the city. Now we kind of have the same design like we had in the aistudio.google.com. We can press F to steal any car. We can move around. We can zoom in and we can zoom out. Press F to get out from any car and have a look. Now, if I scroll down, I did the exact same thing with AI studio, make it to the redesign game at controllable Zoom. You have to ask Demini step by step if you want any changes. Don't make your changes in one single prompt. Build your game step by step. Now, as you can see, we have the weird design for car. So I did not fix the car design because I really like it because I never saw this kind of car anywhere in any game and have a look. We can press F to get out from the car and now check this out. Now, what I did next, I asked Gemni add more depth, add simple character animation, add more life to the game. So right now my game is feel like empty, just me and some cars, that's it. And have a look. We have this character animation. We also have some other peoples in the game. We can press F to enter any car, and now we have also headlights for the cars as well. Now, as you can see, when I drive near to any person, they get scared because we also have some animation. If you hurt anyone, we have the animation for that. So I'm going to show you what I mean and check this out. If I zoom in, we have this red, which means we hurt some people. Now, with the exact same prompt in Google AI Studio, if your character stuck in somewhere or if you are facing an error, you can also ask Gemini. As you can see, my character was not moving, so I ask Gemini character not moving fixed error, and Gemini fixed my error. After that, I ask Gemini add more depth, add simple character animation and add more life to the game. Now, as you can see, guys, we have these NPCs in the game. We have more card in the game, but we don't have any Zoom in and Zoom on feature right now in the game. Now, next, I asked Gemini add aeroplane in game because in real GT one game, we don't have any plane, I believe. So I ask Gemini to add aeroplanes in game, add more depth, detail city, detailed person, and animation. And if we go to the next version, have a look. We have refine animation, we have aeroplanes. We have more life in the game. Have a look. We have more detailed city. Check this out. Isn't this amazing? With a single prompt, we are doing this kind of work, and now we are here in AI Studio. Add aeroplanes in game and add more depth detailed city, detailed person and animation and now have a look. We have these planes added in the game, and we have more detailed city. We have detailed animation for the person. Now next, I ask Gemini make our more functional redesign cars and add more detailed animation in cars and have a look. We have detailed car animation. We have detailed car as well. Check this out. If you hurt anyone, we also have the star for wanted, but we don't have any police cheers. So you can also ask Gemini to do that because in my final version, I did not ask Gemini to do that. Now, if we go to AI Studio, make more functional car redesign cars, add more detail animation in cars. And now if I press start, check this out. We have more detailed car now we have different kind of cars. We can press F to enter any car. We can drive any car that we want. And we also have cars according to the speed limits. So if I get out from this car, we can enter into any car. We have different variety of cars as well. So now if I steal this one, as you can see, guys, this car drive in a different speed, and if I steal the red one, this car drive in a different speed. Check this out. It's like a Ferrari in this version of my game. Next, I ask for add weapons in game, add flable plane. So I want to fly planes because in the real DT one game, you can't really fly any plane. After I ask for weapons and flable plane, we got this error, but we can simply click onto this button fix error to fix this problem. Now, after we fix the error, this is the updated version. You can see, we have this plane, we can press F, and we got this error. Now, IIX Gemini remove weapon from hand, place weapon in random location, also make plane fly in air. Now, after we fix the error and after we remove weapon from hand and place it in random location, this is the version that we got, and as you can see, I find this weapon, so I can shoot it with Control, with control key, I can shoot this person. I can shoot any car if we want. Check this out. We also have destruction animation for the car, as well. Now we can find any plane because I also ask for flable planes. Now I'm going to press F and check this out. We have this fly plane, but the plane is not flying. It's just look like a car driving but with the plane body. Now in AI Studio, if I scroll down, as you can see, add weapon in game, add flable planes, remove weapon from hand place weapon in random location. Also make plane fly in air. And now if I start my game, we don't have the weapon in the hand, let's find a plane, and I believe we have this plane right here. So I'm going to press F inter into this plane and check this out. We have this flable plane. So I really like the design of the AI studio. Believe the AI studio are making more accurate GT one clone than the Gemini. But you can use both because both are giving me different outputs. But with the same controls, now we can land this plane. We can press F to get out from this plane as well. But we don't have any weapon right here. So as you can see, we have this weapon. We can press F. We can't really hurt anyone in the game. We can't really destroy any car right now. Now, next, I polish the design, so I now add more shading and polished graphics. And after this check this out, my game is lagging because we have more shadow, more depth. But as soon as I zoom in all the way in, the game is now running smoothly. Now in the polish version, we have the same thing. We can't really hurt anyone. We can't really destroy anything. Now if I scroll down, I now add more shading and polish graphics. Now may game locate in Paris. And if I go to this version, this is my final version of GT one Clone game. Check this out. We have the Ferrari car in the game. We can press F. We can press F to inter plane. Check this out. Now we are flying this plane. Isn't this amazing the real GTO game, you can't really do that, but I add this feature. And we also have the Paris as well. You can locate your game into any city anywhere you want. Like in Gemini, I asked my game to locate in India, and this is my final version. Check this out. We have our game in India. We can drive this car. We also have COO check this out. Now, you can do so many things in this game. So inspired for my GT one clone, I built this game. Located in India, you can pet any animal. So I added cow, cat, and dog, so we can press E to pet the cow. We can press E to pet the dog. We can move around. We can use scroll to zoom in and zoom out. We can also press E to drive this ikiaPress E to get out from here. You can also press E to drive this car as well. If I press E, get out from the car, if I use the shift, as you can see, we can run fast. So you can also build your own game if you want. And these are all my prompts that I used to build this game. You have to build your game from scratch. If I go to my first version of this game, so this is my first version and I use this prompt, build a low poly three game, a character in a street of India, World can be Explore. And as you can see, guys, without any animation, we have this basic Explore World game. Next, I make it more functional at camera control with mouse. I can control my person with my mouse. Next, I ask for polish and redesign game, add more shading and life to game, add more animation for character and add more detail in this world. Now let's go to the next version. Click here and check out. We have animation for this person. We have detailed CD than before. Now, next, I ask for add more card people, animal, et cetera in the game. And we have next, I ask for add animation for people and animals, add words and play in game, add interaction option to pet animal handshakes with people drive car and buy. So as you can see, guys, I'm building my game step by step. We can press E to drive any car. So I have to find a car and now have a look. We are driving. We can press E to get out from this car. Let's find a NPC. We can press E to do a handshake. Check this out. Let's find this cow. We can press E to interact with this cow. Now, next, I ask for redesign this game in two d vector style, add shading and polish game like a A game and have a look. We have that design that I showed you before. We also have these bikes so we can drive the bites as well. Next, I ask for Mx city design, design, house detail design, add animation for animal s. And if we go to the next version, check this out. We have animal animation. We have dog animation, people's animation, we have house design as well. Next, I ask for add Zoom in Zoom out Control, MP and animal interaction, and more house design. This is my final version that I showed you before. So this is how we can build a clone game from scratch using Gemini three P. 46. Straight Line Motion Graphics: Welcome back, everyone. Now I'm going to show you how we can build this beautiful motion graphics where we can input text. We have customizable Zoom in Zoom out. So let me show you how I built this step by step. We start from here, build simulation ellipse white color star from start of a curve line in S shape to endpoint of ellipse gradient background. If we go to preview, this is what we got from the first prompt. Now in my next round, I say camera follow Ellipse, also at four stop for ellipse for two second a circle for those stop. So in my next version, as you can see, we have four stop, we have second, third, and we have last fourth stop. Now next I ask for add a handheld camera motion, also for stops, input text and make stop circle glow, change shape line to straight line. So I change shape line to a straight line. So this is the version that we got, and have a look, we can input text. We can change any text have a look, and this is real time changing right here on the screen, and this is going on in a loop. Now next I ask for add light leaks and cinematic particle dust effect, add depth of field. And have a look we have particle effect depth of field, and we have cinematic dust effect. Now we also have light leaks as well. Now next I ask add zoomin and out controls, also add fog effect. If we go to the next version, we have zoomin and zoom out effect. Next, I say, make animation smooth. If we go to the next version, have a look. We have this smooth animation. And at the end, I say, add a shock wave circle effect when make a stop point. And in the final version, as you can see, when star a stop point, we have this shockwave effect. Now next you can ask for customize color control. So let's change customized Bclor control, at color controls. Let's send this command. Now next we ask for at color controls, and as you can see, we have the color controls, we can change the star color, and we can change the color of the stop as well. Have a look. So this is how you can build this motion graphics using Gemini three pro. 47. Customize Advanced Slideshow: Welcome back everyone. Now in this lecture, I'm going to show you how we can build this awesome slide show Motion Graphics using Gemini three P. So I have three version. So in my first version, this is where I start. Go to gemini.google.com. If you want to build your app games or motion graphics, simply click here and select Canvas. You can type your prom to build your app games or motion graphics. Now, first I built a simulation, a picture shown floating camera pan left side after two second second image shown, grid background. So if we go to the preview, if I replay simulation, this is the basic simulation that we got. Now next I say polish design add input for images. So I want to change the images. I want customizable images. So if we go to the next version, check this out. This is the first image. This is the second image. We can click here to choose any image from the computer. Now I have two images. If we replay sequence, this is my first image, and this is my second image. Now next I say add handheld camera motion. If we go to the next version, we have handheld camera motion. Now we can replay sequence and have a look. This is a basic sequence. Now we have the second version. So go to astudogogle.com, and by the way, you can build your app games or motion graphics for completely free by using this website. You can go to build. You can click here to hide Navigation Menu. You can click here to show Navigation Menu. We are using Gemini Repro preview model. You can describe your idea and you can build your app games or motion graphics. As you can see, I have this prom, build a simulation, a picture shown floating camera pan left side after two second image shown grid background. The same prom that we use in Gemini, and if we replay the simulation, this is the first image, and this is the second image. Now next, I say, add input for images, add handheld camera motion, cinematic glow, and DC field. And if we click here, we can go to the second version and have a look. Now if you notice in this suggestion in AI feature, we have the suggestion like add image count input, enhanced camera motion control as well. So I selected one suggestion which have dynamic camera transition between state. So if I go to the next version, this is a dynamic camera transition from one image to the second image. So in my next round, I added another AI feature, which is ambient light to the scene that effect all objects. And if I reload this version, as you can see, we got this ambient light effect going on. Now next I say add option to X for images like how many images I want, also add input for images from computer. So this prompt did not work, so I used the ASdution feature, and it gives me this prompt. Allow user to upload images from their computer for the floating frames instead of just providing URLs. Now you can type any changes you want. If you select any suggestion, it will give you a preset prompt. By the way, these preset apps work very well in your apps. It improve your apps a lot. Now in this version, we also have a feature to play the sequence and to stop the sequence. If you go to Config Simulation, you can upload your images. You can also increase the total frame that you want in your image. And let's say if you click here play, let's close Config, look, we have this simulation going on. Now next I say add color controls. And now if we go to the conflict simulation, as you can see, we can change the color of the background. We can also change the saturation of the background. You can click A. You can add your customized color if you want. And we can also increase the frame, upload the image from the computer. So I have four frame. Let's close config, and let's play sequence. This is my first image, by the way. This is the second one. This is the third one, and we have the last image, which is this one. And the sequence start from the starting point. We can stop the sequence. We can also change the full screen, play the sequence. You can record your screen, and you can add this motion graphics in your content. Now for the second time, I use my same prompt in AI Studio and I got an error. As you can see, the screen went black. We don't have anything you are facing any error, autofix button will appear right here, a text that says autofix, you can click on it and it will fix your error. If the text does not appear, you can manually ask Gemni to fix your error. For example, if your screen went black, you can say green wen black, fix this, and that's it. You can send this pom and gemni will fix your problem. Now if we go to the next version where we have the problem fixed, as you can see, guys, we have this slideshow animation. In the second time, we are getting a different slide show animation motion graphics. Now next I say polish design add input for images, and now as you can see, fault, you will have this window where we can input the images. I can click here to upload my images. Now we can click here and now we have the upload file button. We can click here to upload the image. You can also click A and you can page your URL. Now if I click A initiate sequence, we have this slideshow animation, which is very awesome. You have to build this in any video editing software, it will take you hours to create this type of motion graphics. Now next, I ask for ad camera handheld motion, also remove text from images. So as you can see, these texts are very annoying, so I want these texts to remove from my app. So let's go to the next version, and let's initiate sequence. And as you can see, we don't have any text on top of these images. Now next I say, remove all heart from interface. As you can see, these are called Hertz that we are seeing. This is recording. These are some audio level et cetera. These are called Hertz. So I want this to remove from my app. So if I go to the next version, let's click a initiate sequence and check this out. We don't have any hertz. We can reset this initiate sequence. Have a look. Now next I say add camera controls, add cinematic low and depth of field make scene more interesting. And have a look, let's upload the images, initial sequence, and have a look. Handheld motion, interesting scene. We also have this camera change effect by using the mouse cursor. If we reset this, as you can see, we can tilt this we can move this around into any direction that we want. Now next, I say, now add a feature to add multiple images, ask me how many images I want, then input images. So now I want to have multiple images and now check this out. Now first, it will ask you how many sequence you want, so if you want four or you want ten. So let's say I want to have four. Let's click her Continue. Now we have four images. Now we can initiate sequence and have a look we have this sequence. We can change the camera angle. Now next I say add particle effect and add light leaks effect, also add control for color. And now have a look. We have color controls. We can select any color, continue, initiate sequence, and have a look. Now next I say, add a certain camera shake effect to the simulation scene. Use the existing handheld animation or create a new one. So basically, I use the AI feature suggestion right here, so I use one of this feature. So as you can see, I selected Shake effect. You can select refine image animation. You can select any of the suggestion that you want. So let me open the next version, and let's select this color this time, continue initial sequence. Have a look. We have a subtle camera shake effect. This is how you can build your customizable motion graphics slideshow using Gemini three pro. 48. Infinite Ball Fall: Welcome back, everyone. Now I'm going to show you how we can build motion graphic design in minutes. If you are a motion graphic designer, you know how much time it took you hours to create a simple design. Now, with the help of Gemini three Pro, we can build any motion graphic design. So I have this example where we have the white dot fall in front of black background. We also have the text. We also have this effect, and when it enters into white background, it changes its color, and we also have another text behind it, and the loop is going on. Now, if we stop this, we can go to Config. We can change the color of the background. We can change the color of the dot if we can also change the text. We can change the color of this as well, firmar blue. Now if we click here, if we start, as you can see, we have this update, we change this color. So the design is customizable. But let me show you how we can build this step by step. So I did not build this in a single prompt. I have several prompts, but don't worry, the prompts are simple. So we can go back to previous version. I start off with this prompt, build an app simulation, start off with a white circle with black background dropping and dot turn into black when white background at the bottom RI and background turn white. So if we go to preview, this is my basic example. This is the base of our motion graphics. Next, I say camera follow the dot and check this out camera is now following the dot. Now, next I say make it slow motion when enter into new background because the dot is dropping way too fast. So I want my dot to have a slow motion when it enters into new background and have a look. Now next I say make circle drop a little slow. Also when enter in a new background, add a water splash and ocean wave effect when enter into new background. So I want a water splash and wave effect of ocean when entered into new background. If we go to the next version, check this out. We have the water splash and ocean wave Next, I say, add and start and stop button at the top right corner. So I want a stop and start button to this position. And if we go to the next version, we have stop, we have start button. So you can see, we are building this motion graphic step by step. We adding a new feature step by step. Now I say now make background a bit longer. So right now the background is way too short, so I want my background a bit longer, and check this out. We have this long background. Now next I say, add text behind these backgrounds. This is the prom data use. And in my next version, check this out. We have this text. Now, after this prompt, it gives me this nanobnana image. So I again, use this same prompt, and then it give me this result. So as you can see, the dot is passing way too fast and we can't really read the text. So next I say when texts come, add a shock wave and a slow speed ramp effect. Now have a look when texts come, we have the slow motioning effect and we also have this ocean wave effect. Now next I say polish design, add glow and make it cinematic and check this out. We got this result. We have glow and we have cinema. Now next now add an option to change the color of the background and the dot. So if we go to the next version, we can stop this. We can go to config. We can change the color of the dot, the background and we can also change the next I say, make text fit in the screen. So as you can see, guys, if we play this mutation, the text is seeding the screen, so it is going outside the screen. So in my next version, we fix this problem. So if you are facing any problem, you can ask Gemini to fix. Now next I say, add a slide camera shake when enter in new background. So if we go to the next version and have a look, we have this shake. We can also keep this, but at the end, I remove the shake. Now, by using these same prompts, I built two different apps using AIstudo. Now you can go to astoggle.com. You can go to build and you can go to start to build your app. If you have any apps that you have created, you can go to your apps. All the apps are going to show right here. As you can see, we have different animation motion graphics, and I have used the same prompt. I did not tweak anything here and there, as you can see, these are the same prompt that we used to build this app. Now, one thing that I really like in astodt Google is that in the suggestion we have these features, we can select Ai and it gives us this so so I use this feature two time in this app. So as you can see, first time, we have this visual splash effect. If I play the simulation, we have this visual splash effect. And next, we have a certain slow motion background color change. Check this out. So I use these two suggestion from Gemini right here. These are the EI suggestion. By Gemini, you can use any of the suggestion. If we refresh the app, as you can see, instead of falling down in a have this bouncy ball effect and we also have the text. We can stop the simulation, go to settings, change the color of the background, change the color of the ball as well. We can click, change the color of the dot, and change the color of the background. Change this to blue, this one, and we can click a done. We can also change the text, so I'm going to change my text. Let's click Done, play the simulation and Hal. Now, if we click A, we have this feature called AI control. We can increase or we can lower the gravity. We can make it bouncy or we can stop them. So I'm going to type, let's make it bounce now have a look. And if we reset, so I'm going to reset my app. Now, if you refresh your app, it will reset all the things. So we have to go to AI Control, increase gravity. We can press Enter. And now, as you can see, the gravity is now increased. We can play this simulation. We can refresh. We can make ball bouncy, press Enter. And as you can see it make ball bouncy we can play the simulation and now have a look. So this is the first version using AI studio, and by using the same prompts that I used to create my previous app, we got this. Now in this version, I also use AI feature at the end of my app. So as you can see, right here, I use this AI suggestion. If we click here, we have this animation. Now we can ask for loop animation when the ball fall. We can reset the app. We can go to setting. We can change the color of the background, the dot, change the text. And right here, we can also change the background color, the dot color, and the text as well. Now this is my highly recommendation use both Gemini dotgoogle.com and use astodo dotgoogle.com. And use same prom to build your motion graphics because sometimes gemini give you the awesome result and other time the AI stud give you the best result. So I can't really tell which one is the best because they both give me awesome results in different ways. So my high recommendation is used both when building your motion graphics. 49. Bee Hive: Welcome back, everyone. Now, I'm going to show you how we can build B hive motion graphic simulation using Gemini. So first version, I have created this simple version, build a simulation of B hive, and they form name phase. Now, in my next prompt, I say, make B's vector and make two D and check this out. So this is the basic idea that I work in Gemini. So if you click, as you can see, the B spread, and if you let go, they form this name. If you hover your mouse, we have this effect going on. From this basic idea, now in AI studio, I use this prompt, built a simulation Bi, make a name Pasan. Then they fly to the right side and make another name. I'm just add option to input text, also at color controls, also at camera controls, camera follow Bhi. Now, as you can see, all these features are now right here in the settings. If we go to the next sequence, so I'm going to type my name Pasan. I'm going to click here. As you can see, guys, if we click a next fly right, as you can see the fly to this position, but we don't have the name. Clearly, we can't really see. We can turn off the camera. We can change the color of the bees. We can also change the color of the background if you want. We also have the speed control, the digital control, and the size of these. Now if I scroll down, use this AI suggestion feature from the Gemini and I use a smoother transition animation for the Bs. And if we go to the next version, have a look, we have smooth animation of this. Now in my next round, I say, make clear text that can be read easily. So if we go to the next version, let's like this version, I'm going to restore to this version. As you can see, we have this mess. There are so many Bs going on in the screen. Now to fix this problem, I have a suggestion in this AI feature that say, you can improve the rendering, so I selected that version. If you select any changes, for example, if you want to add B trail, you can select that and it will give you a preset prom. You can use it. It will improve your app. I just did that and I'm going to show you my next version. Let's restore to that version. Now as you can see, guys, my app is now running smooth now in this version, I also selected a suggestion feature from the Gemini that say you can form clear text by using the bees. And as you can see, guys, we have this text form, so I can increase the speed. We can load size of the bee. Now we can clearly see the name. Now if I scroll down, I say polish design, add grain retro style, and now have a look. We have this version. You can't really see your text, you can scale down the Bs. Let's scale it down and have a look. Now we have the name we can clearly see. Now I can type any text, send this, go to next step and have a look. The B form that name. Now I can type any text, send it next step, and B form that name. Now you can use this motion graphics in your AI content creation videos because this is a customizable motion graphics. You can change the color of the Bes if you want. You can make them red. You can also change the color of the background as well. 50. Arrow and Sphere: Welcome back, everyone. Now I'm going to show you how we can build this motion graphics using Gemini three. So I'm going to refresh my app. Now, if you want to build app games and motion graphics, simply go to tools, go to gemini do google.com and select Canvas, and you can type your prompt to build your app game or motion graphics. Now I'm going to show you how I step by step build the motion graphics. So first, as you can see, guys, I have this problem to the motion graphic simulation, an arrow going right side, and first a spare come name start arrow touch it, and it shattered into piece gradient background as fair shattered a glass break, sound effect play. And if I run this simulation, as you can see, we have this basic simulation. Now, here's the best thing if you want to build your motion graphics very easily. Have to build your motion graphic step by step. Don't rush and type everything into one cell prompt. Build your motion graphics step by step. Now next I say camera follow arrow, adjust camera view angle. If we go to the next version, if I run this simulation, camera follow the error, we have this glass shatter effect. Now in my next prompt, I say add slow motion effect just before arrow hit sphere. After it hits square, speed goes back to normal. If we go to the next version, I run the simulation, check this out slow motion effect. When the square shatter, we have the normal speed. Now if I scroll down, this is my final prompt. I say add four more sphere, second name, third name, fourth name, and fifth name, change each sphere background as separate level. And if I go to the next version, which is my final version, and now have a look, we have the slow motion effect. We have the first, we have the second sphere. Now you can also ask Gemini for text input in this sphere. You can also ask Gemini for polish design. This way, you will get better result. Now I use these prompt and I built this simulation using AI Studio. Now you can build your simulation apps games for completely free if you use aistudiodgoogle.com. Have to go to build and you can type any prompt right here. As you can see, the model we are using is Gemini three preview. Now, if you have created any apps, your apps are going to show right here in your apps. Now this is the first prompt that we use in Gemini, and this is the simulation that we got. As you can see, we have the exact same simulation. Just a slight change interface and the design. Now we got the next prom, where I say camera follow arrow or just camera view angle. We have to click here and we have to open the previous history. This is your first prom. This is your second, third, fourth, et cetera. So I'm going to open my second version and have a look the camera follow the arrow. Next, we have this prompt where we added a slow motion effect when arrow hit sphere. And if I play this simulation, check this out, we have slow motion effect. Next, I say, add for Msphere the same prom that we used in Gemini, and I'm going to open the fourth version of this app. And if I play this simulation, check this out. We also have background color change when we destroy a sphere. I take this to another level, so I type this problem. Now add full screen mode, also add input text option for sphere. So now I add option for customizable text. So we can type any text that we want. And now take this out. This is the first sphere. This is second, third, fourth and final, which is five. So we can type one. Two, three, four and five. And if we play this simulation, check this out. We have number one. We have two. And we also have this full screen mode. Now you can record your screen and you can edit your YouTube videos or any content creation video, and you can use this motion graphics in your videos. Now next I say add camera control, also add particle and polish design. And have a look. We have the polish design. We can zoom out we can zoom all the way in. We can zoom out, zoom in. We have full screen, check this out. We have this completely customizable motion graphic design. Now if I scroll down, as you can see, guys, I use Gemini's suggestion feature and I add particle rendering logic into my app. So you can also use these suggestion. These are very useful suggestion that we got from Gemini. You can use any of the suggestion right here. Now if I play the app, check this out. We have particle rendering. Have a look. So this is how you can build this motion graphics using Gemini three pro. 51. Pyramid Levels: Welcome back, everyone. Now I'm going to show you how we can build this Pyamed level motion graphics using Gemini three. So I am using astodo.google.com and with simple prompt, build a simulation Pyramid design camera start from top and move toward the end camera stop at each level for 2 seconds, add text for each level, make five level camera focus only on the current level at the time. If we run this simulation, as you can see, guys, the camera focus on each level for two second. We have the description for that level and the animation reset, and we have the whole pyramid. Now I improve this animation by selecting the AI feature suggestion from the gemini. So as you can see, I have selected a suggestion that improve my camera transition animation, and I'm going to show you the second version. And now, as you can see, in this version, we have this smooth camera transition. We can stop this. Now next I selected a suggestion that improve the subtle gradient of the pyramid phases and give them more depth and three D effect. So let's go to the third version. And as you can see, guys, we have this improved design. If we run this simulation, have a look. Now, next, I again select a suggestion that display the title of the currently active pyramid level directly on the pyramid itself. Now let's go to that version, which is the fourth version. We can restart that version. And if we run the simulation, as you can see now we have the name of the current level. Now let's stop this. And next, I also select the AS suggestion. And as you can see, now in this suggestion, we have a subtle Her effect on each pyramid level. So basically, you can select any suggestion, you can simply click onto that suggestion, and it will give you a preset prom. You can use this prom. You can send it and you can improve with a subtle Her effect, check this out. We can select any level, stop this. Now next, I have selected a addition that implements smoother transition. And after that, I add text input for all the level. So I'm going to select that version, and now check this out. We have the description. We have the level. We can change level names, level description as well. Let's start this, and as you can see, we can customize level names. So as you can see right here, we have the description. If we start dissimulation, we have this annoying window appear. Want to remove this window, so I type polished design, ad particle, light leaks, ad color controls, and camera controls. So after that, I'm going to show you, first of all, this version. And now, as you can see, we have the Zoom in Zoom out, we have vertical camera motion. And if I play this, we have this improved design. Now, I was talking about this annoying window. So this time, what I say, remove level description after I say remove level description, it removes the description, but we still have that annoying window. So next I type remove key component window. And if we go to that version, let's run the simulation. And as you can see, we still have this window. So next I say, remove simulation active window because it is the active window. If I open this, this is the active window. I remove this, and now if we run this simulation, we don't have that window. Next, I selected a suggestion that modified the Her effect, and next, I say, remove hertz when simulations start. So I want to remove these hertz when the simulations start. So let's go to that version. Now if we start the simulation, we don't have any hertz. Next, I selected a suggestion. I never clicking on a pyramid level to set it as the active level. And after that, we have this suggestion. So basically, I have selected several suggestions to improve my app. You can do that as well to improve your motion graphics as well. Now this is the latest version, and now have a look we have these levels. We have camera controls. And if I start my simulation, this is the finest simulation that we got. So it will give you the suggestion according to your design, according to your app and game. You can select any suggestion and it will improve your app. 52. Clock Motion Graphic: Welcome back, everyone. Now I'm going to show you how we can make clock motion graphics animation. So if we go to the preview, this is the basic motion graphics. This is my prom built simulation, a ellipse, white color, start from start of a clock to the endpoint of clock gradient background. So this is basically a customizable clock animation. We can increase the trail. We can lower the frame of the trail. We can also increase the speed of the or we can lower you can enable or disable the grid. You can enable or disable the next I say camera follow Ellipse, also add four so for Ellipse, for two second, add circle for those stop. Let's go to this version and have a look. We have four stop. We can increase the orbit speed, check this out. We can pause it, we can reset this. Now next, I as four add a handheld camera motion, also for stop, add input text and make stop circle glow. Let's go to the third version, which is this one. And as you can see, we have the text input. We can input any text we want. So this is a customizable motion graph. If I scroll down next I ask for add light leaks and cinematic particle dust effect, add depth of field. This is the fourth version, so I'm going to select my fourth version. Let's restore to this version and check this out particle effect and lightly. You can also increase the size of this clock. Next, I ask for add zoom in and out control, also add fog effect. Also make animion smooth, add a shock wave circle effect when make a stop point, also add color control. So let's go to the fifth version of this app and as you can see, we have this design. So we have this error because this is the fog, this is not an error. So next, I ask for polished design, make it cinematic. And if I go to the sixth version of this app, you can clearly see we have cinematic app design. We can zoom in zoom out. This is the new feature, and we can change the color. Of we also have the time deletion, which is the orbit speed. We also have camera. We have fixed camera. We have camera flow. We have hertz, we have orbit, and we have the text. Now, next, I say remove fog because this is a fog. So in my sabbath version, I remove the fog and we have this precise version. We can change its color. This is a fully customizable motion graphics. Now next I say poise design add particle and add le and this is my finished design. Have a look, we can increase this. We can change the color, begin Zoom and have. Next, you can also ask for hide the menu. So let's send this prompt. Now as you can see, we ask for Hide menu option, and if we click here, we can hide the menu. Now we can view this in full screen. You can record your screen and you can use this motion graphics in your video. Now, next, I select one of the AI feature suggestion from the gemini and I enhance the shock wave, so I'm going to show you that version. And now as you can see, we have this double shock wave effect. Next, we have a suted rotation effect, implement a very slow sued rotation effect suggestion from the gemini and if we go to that version, check this out right here, we have this rotation. If I disable my camera follow effect, now we have this auto rotator. So all of these editions are very great. You can implement all of the suggestion that Gemini suggests you from this AI feature. If you enable the follow, it will follow this dot. Now, next, again, I use the suggestion and I add some dust particle movement logic into my scene using the Gemini AI suggestion feature. Now as you can see, we have these particles in the background. Now, next in my background, I add Siler flares. Let's open the tune. Now in the background, if you notice, we have Siler flares. So let's hide the menu so we can clearly see. So these are called StelarFlare. Now, next, I change my style to retrostyle and this is what we got, so I did not like this style. So in my next prompt, I remove the retro style, and after that, I add dynamic lighting effect, and at the end, I use, again, the AI features. I'm still using the AI features. I add some dust particle movement logic into my let me show you the final version of my app. We have this improved app. We have simulation speed. We can stop this. We can hide the menu. We can also click here to change the colors. So this is very impressive. We also have this auto gyro effect. Let's play the simulation and now have a look. Let's view this in full screen and check this out. So this is how we can build this motion graphics from scratch using Gemini three.