Make a Realistic AI Avatar Clone with ElevenLabs (Full Transcript)

[00:00:00] Speaker 1: Imagine you could create a custom AI avatar that looks so real that people didn't actually know it was you. Well, you can. What you just saw was a digital clone of myself that looks and sounds 100% like me. And if you didn't notice, it's because I created it using the Eleven Labs creative platform. There's a lot of different AI tools out there nowadays, but when it comes to creating AI avatars and cloning yourself, the quality is the most important thing. And thankfully, Eleven Labs provides all of the best AI image and video models out there and voice clone technology, allowing you to make the most realistic AI avatar clones out there.

[00:00:28] Speaker 2: What do you think? Believe it or not, I'm also an avatar from Eleven Labs.

[00:00:33] Speaker 1: And now, if I put myself side by side with my AI avatar and I lip sync the audio, can you really spot the difference? And the cool thing is, is that you can even change the clothes and put yourself in any environment. So I decided to make myself look much smarter than I actually am. Hey everyone, my name is Alec, and today I'm about to show you how to get the most realistic clone of yourself. You can simply type out what you want on your keyboard and create your video without having to go through the painful process of recording. So every take from now on, it's your turn. It's going to be perfect. Let's begin. Now with Eleven Labs, you can create avatars that look 100% like you, or you can make entirely AI generated avatars, and you can even do a mix. So you can put yourself in any environment. But first, let's take a look at how to clone your voice so you can create voiceovers that sound like you without actually having to record them. To clone your voice, you need to click on Voices, and then on the right, we want to click on Create a Voice Clone. And here we have multiple options to clone our voice. First of all, we've got Instant Voice Clone, which allows us to clone our voice with a minimum of 10 seconds of audio, and then we have Professional Voice Clone, which allows us to create a super realistic digital replica of our voice. And here we need to submit a minimum of 30 minutes of audio. And now the higher the quality of the audio you submit to your voice clones, the better the result of the voice clone will be. And so if we click on Professional Voice Clone, as you can see here, you simply want to click Create a New Clone and now go through the process of uploading your sample. You can even record directly into Eleven Labs and then name your voice clone. And if you want an in-depth guide to creating the highest quality voice clone possible, you can click on Create a New Clone. We'll put a link down in the description below to a full 10 minute guide that we've put together for you. Once you have your voice clone, we can then click on Text to Speech. And here we can type out anything that we want our AI avatar to say. And so here I've got a default script talking about coffee. But to use our voice clone, we can then click on Voices and here we can go ahead and search for our voice clone. But as you can see here, I've got my own professional voice clone that I've named Alex PVC. And so if I click on it, I can now generate this text right here that I've typed in, and it will sound just like this. We can then click on Create a New Clone, and here we can see that it sounds just like me. So let's take a listen. Coffee is one of those things most of us consume every day without really thinking about it. And so as you can see, I've now got a script that sounds just like me, and all I've done is typed it out. And so I could go in and replace this with anything I like. And so if I said, Hello YouTube, please subscribe, we can then click on Generate. Hello YouTube, please subscribe. And as you can see, we've now got Hello YouTube, please subscribe in my voice. And so now that you have your voice clone and your voiceover, we actually need to make the AI avatar. And to do this, we can head to the image and video tool on the left right here. And with inside image and video, there's a bunch of different ways that we can use and create AI avatars. We can use default avatars, we can create our own from scratch, but we could also upload pictures of ourselves and even customize those avatars to get the exact result that we want. And so to begin, let's start with the default model. If we switch to video in the generation model, we can then click on the video models and we can scroll down and we can select Creatify Aurora. Within Creatify Aurora, we have a bunch of default avatars that we can choose. And so let's say we decide to choose this one right here, I could then click on speech. And here we can go and add the audio that we've just generated. So if I click the plus I can use this as the speech. And now I can simply click Generate and Creatify Aurora is then going to create an AI avatar video of the default avatar plus the audio that I've just given it, which is my voice clone. But if you can't find an AI avatar you like, you can then actually go and create your own simply by prompting the avatar that you want in the image generation model with any image generation model. And so we could select a model like Nano Banana Pro. Here, I'm going to go ahead and then use this prompt. And this prompt is just to give you a little bit of inspiration. But you could go ahead and describe any AI avatar for any use case that you like. So it could be for a construction site, it could be for a scientist, it could be for a doctor, and you simply type out what you want. Here, I want to change the resolution to at least 2K, but I would recommend even going up to 4K because the better the input of the image we give to AI video generation models, the better the output. And then I'm going to choose four generations just so I've got a choice to choose from. And then I simply click Generate. And as you can see, now we've got four different AI avatars that we could choose from. And if you didn't like any of them, you could go and click on the prompt again, regenerate to iterate on top of it. But if you did, what we could do is switch back to video. And then within the Creatify Aurora model, we could simply drag and drop the one that we like into the AI avatar space. And we could scroll down to our previous generation and simply click on the audio again to use the same one, and then click Generate. But for this use case, I want to make an AI avatar that looks and sounds 100% like me. We already have the voice, so now all we need is a photo and an image of me. Now, if I was to click on Avatar and then I click on Upload, I could open up an image of me and then simply click Generate. And now I have an AI avatar video that looks and sounds like me. But the problem with this video is that the image is one by one. And I also don't like the layout on the backdrop. It doesn't feel like it was recorded and shot. It looks like we've turned an image into a video. And so what we could do if we go back to the image generation model, I can paste in this prompt this time. And so as you can see at the start, I've got use my likeness from the uploaded reference image. And then I'm going to upload that same image as the reference image with Nano Banana Pro. And then I'm going to change the aspect ratio to 4K because again, I want a 3D input for the AI video models. And now I've got a short description of what I want the setting to be like. So I'm putting myself in a new environment and here I could change my clothes. I could change the background. I can change the lighting, anything I like. And by the way, let me know in the comments down below what type of AI avatar you are making. I'm very curious and I'd love to see what you guys make. And before I click Generate, I do want to mention that you can adjust the aspect ratio. And so here I want to turn it into a 16 by 9 video, but we could also make this a 9 by 16 video. So if you were making a vertical short form Instagram Reel or TikTok, this would be perfect. And if you want something a little bit more UGC looking, we actually have a full tutorial on how to make UGC videos with AI. And we'll leave that in the description down below as well. And so now that I've got my reference image and the prompt, we simply click Generate. And so as you can see now, I've got four still images that look 100% like me that I could use as my AI avatar. And so I really like this one. The environment looks so much better than the previous one. And it looks like a YouTube video, which is the aesthetic I was going for. And so then again, I could go back to Creatify Aurora and I could drag and drop this one in as the AI avatar image that I want to use. And then again, go and add the speech simply by clicking on the previous audio that I used. And I would then go ahead and click Generate. And now I've got a custom AI avatar that still looks 100% like me and still sounds 100% like me, I'm in an environment that I was never in. And right now I've been using Creatify Aurora for all of my generations. But I do want to talk about a couple of other models and their limitations to help you pick which one you can use. Because if you click here, as you can see, there's a bunch of different AI models that you can choose. And we've been using some from the lip sync category. But there's three that I want to talk about. The one that I've been using for all the generations so far has been Creatify Aurora. And the reason I like this model is that it has a maximum length of 60 seconds. However, the resolution is to a maximum of 720p. But it's very expressive, and it's pretty good for stable talking avatars. The minus of it is just that you always need to upscale it afterwards, which we can also do inside of EleveLabs as well. The next one is OmniHuman 1.5. This one's nice, it's got a maximum resolution of 1080p. But the duration is capped at 30 seconds. It is really good at realistic faces, though, but the clips are a little bit shorter, and there's a little less movement and expressiveness in the face. And then finally, we've got LTX Audio to Video, which recently came out. And this one, you can do resolutions of 720p and 1080p. But the maximum length is 20 seconds. However, this is an audio driven AI video model. And so it's not strictly lip sync. This means that you can add a custom prompt with it to guide what's happening inside of the video. And it will take into account your prompt and the audio, as well as the AI avatar start frame that you've given it. And you can get some really cool results with this. And if you think about it, there are so many different ways that you can use this AI avatar content creation workflow in your creative processes. And let me show you a couple of cool things. First of all, if you didn't want to create a voice clone, you could actually upload your own voice recording. And so you're essentially lip syncing a real audio recording of your voice to an AI avatar. And that's one workflow that you could do. But then if you think about it, you could take it even further. And here you could go to voice changer, you could then record your own audio directly into voice changer. And you could then swap it to a new voice. So here right now, I'm recording. And then if I click stop, I can then go to the voices section and change it to any voice I like and get a completely new sounding version of the recording I've just done and lip sync that with an invented AI character to get some really cool results.

[00:09:44] Speaker 3: That is just a few of the many ways you can use the Eleven Labs creative platform.

[00:09:50] Speaker 1: And now you know how to create AI avatars with Eleven Labs that look and sound 100% like you. If you have any questions, let us know in the comment section down below. And if you enjoyed this video, please hit that like button and don't forget to subscribe. Thanks for watching.