Make videos with AI in ElevenCreative
Chapters15
Introduction to the stream and the goal of creating video content with AI using 11 Creative.
A hands-on walkthrough of making AI-powered videos with ElevenCreative, showing how to storyboard with images, turn scenes into videos, and stitch it all together using flows, studio, and GPT Image 2.
Summary
Alec from ElevenLabs introduces Eleven Creative and demonstrates building an AI-driven video from scratch. He highlights the latest GPT Image 2 release and explains how to combine image, video, audio (text-to-speech with 11v3), and music in a single workflow. The session walks through generating static hedgehog scenes, turning them into video prompts with specific models like video 3.1 fast and cling 3.0, and using character reference sheets to keep the avatar consistent across scenes. Alec emphasizes storyboarding with AI by starting from static images and then converting frames into videos, all while maintaining consistency with reference sheets saved in the Files system. He also introduces Flow, an infinite canvas for connecting image, video, TTS, and music nodes, plus a composition node to assemble multi-scene timelines. The live demo covers voice delivery using audio tags in 11v3, adds background music from 11 Music, and shows how to sync audio with video, trim clips, and adjust fades for a polished end product. Finally, Alec points to Studio as a more traditional editor that still leverages AI, and encourages creators to experiment with ads, shorts, and even TV-show scale content using these tools. The video closes with a teaser about 5,000 free GPT Image 2 credits and a note that the replay will be available on YouTube for on-demand viewing.
Key Takeaways
- Use GPT Image 2 inside 11 Creative to generate high-quality hedgehog visuals, then convert chosen frames into video with the video generator (e.g., video 3.1 fast).
- Create a character reference sheet with GPT Image 2 and tag it in your prompts to maintain consistent character design across multiple scenes.
- Flows automates the end-to-end AI production: image generation -> video generation -> TTS -> music, all connected as nodes on an infinite canvas.
- Audio tags in 11v3 text-to-speech let you control delivery (whisper, pause, affectionate tone) for more expressive narrations.
- Use the composition node in Flows to build a quick timeline of multiple scenes, then run to render the full video.
- Music generation in 11 Creative (11 Music) supports instrumental prompts and emotional cues, and can be mixed as a separate audio track in the timeline.
- Studio provides hands-on editing with all AI assets carried over from Flows, enabling precise trimming, sequencing, and lip-sync when needed.
Who Is This For?
Ideal for creators who want to scale AI-generated video production: YouTubers, social video teams, and indie filmmakers curious about AI-powered storyboarding, consistent character design, and end-to-end automation.
Notable Quotes
"'GPT Image 2 was just released as well yesterday. You can already try that inside of 11 Creative.'"
—Alec introduces the GPT Image 2 release and its integration into 11 Creative.
""Flow is an incredible new way to create. It's an infinite canvas that allows you to connect and automate the workflows.""
—Explains the purpose and power of Flows for automating AI production.
""Audio tags allow you to guide and direct the delivery of your voiceovers to perfection.""
—Demonstrates how to fine-tune TTS delivery with 11v3.
""Lip sync is 100% possible.""
—Direct answer to a live question about lip-sync capability.
""We can drop in a character reference sheet and use it across all scenes to keep the hedgehog consistent.""
—Shows how to maintain character consistency with reference sheets.
Questions This Video Answers
- How do I use GPT Image 2 inside Eleven Creative to create video from sketches?
- Can Flow automate the entire AI video production pipeline in Eleven Creative?
- What are audio tags in 11v3 and how do they affect voice-over delivery?
- How can I maintain character consistency across multiple AI-generated scenes?
- Is lip-sync supported in Eleven Creative's AI tools and workflows?
Eleven CreativeGPT Image 2Flow (Eleven Labs)Image and Video generationText to Speech 11v3Audio tags11 MusicCharacter reference sheetCling 3.0Video production workflow
Full Transcript
Heat. Heat. Heat. All right, it looks like we are live. Excellent. Okay, we've got a few people coming in. Thank you for waiting through that five minute countdown. Awesome. Hey everyone. Um, my name is Alec. I do social media 11 Labs. Today we're going to be talking about and I'm going to be showing you how to create videos with AI. And so we've done a little bit of a uh a countdown. So I think we might just get straight into it. So, um, first of all, yeah, actually drop in the chat where people are coming from.
Um, we've got people coming in from Santa Clara. That's awesome. There's quite a few people viewing. We're streaming from to multiple platforms today. We got people coming from Thailand. Excellent. Okay. So, today we're going to be using 11 Creative and I'm going to walk you through the process step by step of how to create a video using AI. And 11 maps has evolved a lot over the past year and in the past few months. And the team is shipping incredible products. And so I want to show you how to use them, combine them all together to create some incredible video content.
And uh GPT Image 2 was just released as well uh yesterday. Um you can already try that inside of 11 Creative. Um and so I'm going to try and tie that into the uh the process today. But uh brilliant. And uh I also want to make this a little bit interactive. So anytime throughout the entire stream, please do ask questions, drop questions in the chat because uh when we generate with AI, there's a little bit of a waiting time. But let's begin. So here I am on 11 Labs and we're going to be focusing specifically on 11 creative which allows us to create uh videos from start to finish.
So that's the audio um whether that's the voice over whether that's the music whether that's the sound effects and then now combined with image and video and the best AI image and video models in the world all in one place. So 11 Creative now really is your go-to spot when it comes to creating content um and especially when it comes to creating content with AI. So, I'm just going to quickly log into my account. And if you are an 11 Labs user, you are very familiar with this, but we've recently updated the dashboard. Um, so you have a quicker and easier look at all of your generations, your recent projects, and we've even recently added tutorials that allow you to to dive deep into specific workflows.
So on the sidebar here on the left, we have a bunch of different tools um for audio, image, and video. And then we also have uh a couple of different interfaces that you can use all of these tools in. And so I'll quickly just break some of these down and then we'll get straight into the actual flow of creating the video. The tools that I want to be using today, first of all, obviously we're going to be using text to speech. the the content that I want to create is a short trailer, an animated style trailer using GPT Image 2, maybe using Cling or maybe some of the the latest models that we've just added like Sea Dance 2.0.
But I would love to create a little trailer start to finish. I've got a little story that's prepared about a little hedgehog. Um, it's a short animated story, but everything that I'm going to teach you today and show you can be applied to any type of content you're making. Whether you are creating short films, whether you're creating uh faceless YouTube content, whether you are creating ads, I want to show you how to use the tool so you can actually go and create any content that you would like to create. So, first of all, obviously, we're going to be using text to speech for uh the voice over.
And I want to make sure we can see this. Let me actually zoom in a little bit here. Text to speech for the voice over. We're going to be uh typing out the voice over. We're going to be using 11v3. I'm going to show you how to use audio tags to get the expressive uh delivery of your voiceovers and how to get the perfect delivery so you can really control uh the intonation and the flow of that voice over so it matches your content perfectly. Then we're going to be using music because I want to generate some uh background music to our video.
And then I want to obviously be using image and video which is where we generate still assets and also videos. And we're making a video but we're going to want to generate some images. and I will explain why uh right now. So, let's begin. Let's start with image and video. So, for those of you that aren't familiar with image and video, image and video is essentially, let me zoom out a little bit more. It's essentially uh an all-in-one image and video generator that combines all of the best AI image and video models in the world all in one place.
So, any asset that you need to create, you simply go down to this little prompt box right here and you type out what you would want to create. So for an example, if we wanted to type out and generate uh a hedgehog, I could literally do so. I type out the prompt below that I can select the model. So here, if I click Whoops. If I click on uh Nano Banana, we've now got all of the best image models. So for an example, let's use uh GPT image 2, which just came out yesterday. You can already try it inside of 11 Creative.
We can now go and uh generate. And so I'm just going to generate. And so while it generates, I'll quickly explain what uh we did. But at the bottom right here, when we generate, there's a bunch of different settings that we can tweak and adjust. So for example, we're generating this static image of a hedgehog. I can choose the aspect ratio. I can choose the resolution. And with the latest GPT image 2 model, you can go all the way up to 4K. But for a lot of the generations I'm going to do throughout this um this Whoops, my phone's going off.
for this uh for this tutorial, this uh live stream, uh I'm going to be generating at a lower resolution just to save time on the generations. After that, with the GPT image too, you can also choose the quality. Then we've got the number of generations. And the last little feature that I'm really loving is um a little toggle that you can switch on that will automatically improve uh short prompts. So, for example, here I've just put in the word hedgehog, but 11 Labs allows me and inside 11 creative, I can toggle this on, and it will essentially enhance this prompt so I get a better looking output.
And so, as you can see, we're about 88% complete with this uh generation of a hedgehog. And here we are. We have now a lovely little picture of a hedgehog. And when it comes to generating video, it's the exact same process except we want to go and click on this little video toggle. And now we click on video and we've got access to the video generator and it works exactly the same. You've got all of the little settings that you can customize and adjust at the bottom and then all of the all of the best video models in the world with uh one of my favorites being cling 3.0 and the uh of course seance 2.0.
And so you can use all of these models and generate all of the videos and images you like inside of 11 creative. Now, for my little story, this is how I would generate uh a first scene. So, I want to create a little animated story of a hedgehog. And so, we're going to go with a specific felt style animation um that's been very trendy on social media recently. And so, I've got a little prompt that I'm just going to grab over here. And I'm making video, but obviously we're going to start with static uh static images.
And you'll understand why in just a second. But here I am going to generate this and then we are going to uh break it down a little bit and I'm going to be using GPT image 2. I'm going to turn off uh automatic restyling or rewriting of my prompt. But here, let me zoom into my prompt. As you can see, this is the the scene that I want to create. So it's a 100% felt made scene of a tiny hedgehog peing out of a pile peeking out of a pile of stitched autumn leaves. I'm describing like the overall scene.
Then what I'm doing after that is I'm adding a bunch of more context and a bunch of more detail to really make my prompts unique. And then at the end, what I like to do is make sure that no text is actually generated on screen. So I include, I guess you could call it negative prompts in the prompt itself. And now, as you can see, we've got this cute looking hedgehog. And the reason we start by generating with an image is that we first want to make sure we find the exact scene or the exact style that we like and then turn that into a video.
So now what I could do, let me get rid of this prompt. Now what I could do is we could toggle straight to video or on the scene we like. Let's say we like this little this little hedgehog right here. We could simply click on the little video button. And now it's taking us straight to the prompt box and we're using this scene as our start frame. And the reason we want to create these static images first is because you can you can look at it as storyboarding. When you're storyboarding, you're generating each scene or you're drawing each scene as a static to then turn it into a video later just to make sure that you get uh every scene correct or every scene right.
The only difference to storyboarding when you're doing this is that we are always creating the first frame to our video. And so we're then going to turn this little uh hedgehog start frame into a video. And so to do that, I'm just grabbing a little prompt off screen right here. And the prompt we're going to be using is uh a tiny felt hedgehog blink slowly. So here we don't have to give it too much context and detail as to what we want to see visually because we've already done that with the image and now we're just turning this into a video.
So we want to describe the action or the animation that we see in the video. And I'm quickly going to change this model to um for the purpose of this because video will take a little bit of time. I'm going to use a quicker model. So we're going to go for let's video 3.1 fast. But you can go and use cling 2.6 or cling 3.0 which is a great model for generating animated content. Um and then what I'm going to do is I'm going to tweak a little bit of the settings. Again I'm going to leave it at a lower resolution just for the sake of things generating quicker throughout this tutorial.
And then I'm going to change the duration to six. Uh I don't want to generate with any sound because I don't need sound. And then let's just generate two different variations. And now I click generate. And now what's going to happen is V3.1 fast is going to take this prompt and this start frame and turn it into a video. So it's going to start on this frame right here that we've chosen and then use the prompt that we've given it where we're describing the actions we want to see to turn it into a video. And while that generates, let's have a look at how you can consistently get the same character throughout the um rest of your content because that's one thing that's very tricky with AI, right?
It's keeping stuff consistent throughout multiple different scenes, multiple different generations. So, what I'm going to do is I'm going to move back to video and we're going to use again this same this same scene because I really like this hedgehog. I want my hedgehog to look like this one every single time. And it's got this little these little spiky spiky felt uh uh spikes on the back. And so I'm going to drag this in and use it as an image reference. And this time um let me see what we're going to I want to create a character reference sheet.
So create a character reference sheet. And what's really cool inside of 11 creative is that you can tag specific references. So here I would say create a character reference sheet from and then if I do the at sign and I'll zoom in a little bit here. here so we can see it. If I do the app, I can then tag the specific image um that I'm talking about. And GPT image 2 is actually really great at creating these layouts. And it also has um real world knowledge. So it will have a lot more context um that it will be able to pull from to create a really good reference sheet.
So I can talk to it as if I was talking to a um a designer or or just another person. So, I simply just describe a natural language just like I would chat with an LLM. So, create a character reference sheet from image one that I can use for consistent um outputs in future videos. And I haven't I haven't tried this one yet first, but um this is the first time adding this on the end, but let's try it. And again, generating at a lower resolution just for speed purposes. But now, I'm going to just generate two.
And we hit generate. And so while that generates, we can have a little look at our video. And so, as you can see, we've generated this cute little hedgehog that is now crawling out of the pile of leaves. And this is the first scene that we uh have created for our video. And what I would do next is I would essentially get my next prompt. And so, let's just do that. For the second scene, I would go and get my next prompt. I'm just grabbing it off screen again. And it's similar in terms of the style.
Um, again, 100% felt made scene of a tiny hedgehog, but this time we're describing a different location. Uh, the different kind of, um, elements and assets within that static image. And then I would generate that. I would go through the same process, but for it to be consistent, I would want to use the reference uh, sheet that we've just generated. So now what's really cool with Jim um with Jimage with GPT image 2 we can create these character reference sheets and so now I have uh all the different angles the different colors um I can see him from the front from the side profile and I can then use this as a reference for every single future generation so I make sure my hedgehog is always the same character.
So, this means that now when I'm generating my second scene here, I can simply drag my hedgehog reference sheet and use it as the reference. And so, in my prompt, when we're saying a felt made scene of a tiny hedgehog, um, we could tag it right here, tag the reference, and then we can click generate. And so now I click generate and we would then get the second static image with the same hedgehog in the new location that we would then use as the start frame to turn into a video. So we're repeating this cycle where now I would need to go and get the prompt that I would like to turn uh or that I'd like to use to turn the static into a video and then kind of rinse and repeat the process.
And so the way we're doing this uh yeah and so the way we're doing this now I've got this second scene we generate another one. So the way we're doing this is very linear. So once again now I would go and click on this. I would click video here. I would go and grab my uh second prompt for the second scene. And this time we've got the hedgehog toddles forward with small careful steps. And once again um I would just click generate. And so we're going through this very uh repetitive process. And this is how you end up creating your video scenes.
And I'm going to get to how to to stitch them together in uh just a little bit. And then we're also going to add the voice over. We're going to also add the background music. But what's really cool is let's say um I want to reuse this character reference sheet obviously for all my future generations. Recently, the 11 Labs team uh released a little feature that is very, very useful called files. And so, it's this little toggle. I believe you can see, yeah, this little toggle on the top right that allows you to create folders and put in all of your favorite assets and generations.
And so, I've already created one right here called Hedgehog Story. And you can literally create one by clicking this little plus icon, hedgehog story. We can go and grab our Hedgehog reference sheet. So, I believe we use this one right here. And I could just drag that and drop that in my file. And I could call it reference sheet just like so. And so now in all of the other tools that I go to inside of 11 creative, I can easily find my reference sheet. And so what I want to use next and what I want to show you is how to use flows.
Flows came out, I believe, just over a month ago. And as you can see, we've got our cute little hedgehog kind of wandering through right here. But flows came out um I believe over a month ago now. Might even be might even be two. I don't know. Time flies. But it's an incredible new way to create. So what we found when creating with image and video and creating AI video content is that it's a new way to create, but we're still using a very linear process. And so with flows, it's an infinite canvas that allows you to connect and automate uh the workflows.
And so the whole process that we've just done of uh creating the prompt, selecting the model, generating the image, turning that into a video, that repetitive process we can automate. And so let's go to flows. Flows right here inside of flows. Um we can simply click on new flow in the top right. And now we are inside of flows. And I'm going to move to a template in just a second to show you what is possible. But this is how flows works. We've got all of the same tools that you can access inside of 11 creative.
So, we've got image and video where we can use all of the best uh image and video models in the world. We've got text to speech with 11 v3 for the most realistic voice in the world. We've got 11 music. We've got sound effects. And we can create individual nodes on this canvas. So, here's how it works. First thing I want to do, and I'm going to go back and grab the the first prompt. So, we used um yeah, we're creating this little hedgehog short story. And so, I can go and create by right-clicking create an image generation node.
Or you can find all of them at the bottom right here. So, I can select image generation node. And just like this, we have this little node, and it looks different, but it works exactly the same as the prompt box. And at the bottom, as you can see, we've got all the models that we can choose and select. And so, these are all the image ones. And so again, let's say I wanted to use uh GPT image 2. We can then adjust and tweak all of these settings. And then I can go and type and describe the prompt in right here.
What we can also do is we can have the prompt as a separate little node. Um, and I like having it this way to kind of visually see everything that goes on. And you'll understand why in just a second, but I'll paste in my prompt. And so here we're building out the flow. And you'll quickly see that we're basically getting a visual representation of everything that I've just done inside of image and video. And so now from this image node, we would turn this into a video. So I can click and drag from the connector.
So you literally just click from this little blue dot right here, drag it, and then we can click on video generation. And so now we've got a video generation node. And once again here, we're going to put in the prompt that I would like to uh use to turn um this image into a video. And again, you can paste it uh in the bottom just like this. Or you can also just do a little uh text node. And let me clean this up just a little bit. Make it nice and clean. I like having them separate.
And again, I'll show you why in just a second. But now we've built out this entire flow. And earlier we saved our hedgehog reference sheet because I want to use the same character throughout all of my generations. So I'm going to open up files. Again, you can access it from all of the different tools. And then I can simply drag my reference sheet in and boom, now we've got this little hedgehog reference sheet that I can see visually. I can place it wherever I like. And so it's just a very useful way to create. And I was talking about storyboarding earlier.
This is actually a very useful um a very useful thing to be able to do is just kind of arrange generations and put them where you want. But the purpose of the reference sheet here is that obviously I'm going to be regenerating the first scene because it turns out I might want to change the location or I might want to um change the environment or add in a few extra assets, but I want the character to stay the same. And because I can drag in my reference sheet, I can then use it as an image reference.
And so this little um connector right here is for reference images for the image generator. And so I can literally just connect it up like so to the reference sheet. And now it's using the text prompt and it's also using um the reference sheet to create the first scene and then it will turn it into a video. And before there was a lot of clicking and dragging and it was a very repetitive process. But here now that we've built out the flow, we can regenerate it as many times as we want without having to go through the whole process.
And what's really cool is I can simply click run from here. And now, as you can see, everything is being generated. As soon as the video, as soon as the image is finished, it will then generate the video. And it will do that automatically. And when you're creating just one scene, this is great. But the power really comes from when you are creating multiple scenes. And we're creating a short animated story here. I want the hedgehog to uh wake up through the pile of leaves and then walk through uh nature. And so this is the flow that I've created for the first scene.
And so if I wanted to then create a second scene, I can literally just select everything right here. So the text prompt for the image, the image generator, the text prompt for the video, and the video generator. I can hold down option on my keyboard now that I've selected all of them. And I can simply drag and drop. And now we are creating uh Whoops, disappeared. But now we are creating uh let me try that one more time. I think it might be because I am in the middle of generating. But now we are creating a second scene.
Okay, we're having a small hiccup there. Let me refresh or actually let me just jump straight to um a different flow. So this this is exactly what we have just um just made right the first shot. Okay. Text prompt, image generation, video prompt, uh, video generation. And all I've done, so holding down option on the keyboard is selecting it, clicking and dragging. And now we're basically creating the flow for a new scene. And so what I want to do for the text prompt is simply swap out the text prompt. So for the new one. So here, as you can see, we've got the second prompt.
So we've got the hedgehog in a new location. We've got the uh video prompt that turns this static image into a video. And then we rinse and repeat that. But every time we're duplicating, so this section, this section, this section, the reference sheet is always staying on as a connector. And so this means that it's the same little character, same little hedgehog throughout the entire creation process. And so now I've got all of my scenes with my consistent character all the way through. And so this is the power of flows and it is such a fun way to create um images and videos and kind of play around with it.
But flows goes much further than just image and video. Like I mentioned, you can find all of the tools inside of 11 creative that are inside 11 creative inside of flows. So the next thing that I might want to do here is start stitching this all together to see what that looks like. And we can do this with a composition node. And so here at the bottom we can go and select our nodes. Um we got image generation node, video generation, we've got text to speech, sound effects, music. Um we can also actually use an LLM node to create prompts with the LLM.
And then we can use what we call a composition node. And so if I click on this composition node, we've got this cool little node right here, which is essentially a mini timeline node built into flows. And this allows us to compile all of the clips that we've uh created together. And so I can simply do that by dragging from the video connector to the composition node each of the scenes. And as you can see, every time I drag one in in the timeline, we're seeing them. Oh, I'm just having a quick issues with my Oops, it's frozen.
There we go. We can see each clip one after the other. And so I simply have to then drag all of my clips. And have I forgot any? Oh yeah, I forgot this one at the beginning. Um, let me let me just redo that. I don't want to forget the first scene. But it then allows you Oh, I should have actually just rearranged it because you can rearrange in the uh in the timeline. Not being very efficient here. Right, there we go. Let's just drag those in. And now they compile into the composition node one generation after the other.
And this is really cool because it very quickly allows you to see at a glance what the flow of your content will look like. And so I've connected up all of these scenes and now I can simply click run and then it will quickly compile those together in literally just a few seconds. This usually doesn't take long at all. And then we can preview what the entire video looks like. But obviously now that we've created the video, we also want some music. We want a voice over. And let's just take a quick little look. As you can see, the hedgehog wakes up, climbs out of the leaves, walks through the meadows and the fields, and then nighttime comes, I believe, or it's the evening, he finds his little friend, the bee.
Um, and we've created this consistent animation that goes throughout the entire day of a hedgehog. And we can see what the entire thing looks like all the way through. Now, the power of nodes is that at the end here, um, we saw, and if I just go over here on my little title screen, we made a mistake. And this mistake was made on purpose just because I wanted to show you the power of nodes and how you can edit and tweak all of your AI visuals in a very efficient way. But we've got a little hedgehog on here, but we didn't plan to have the hedgehog uh in this generation, and the prompt didn't plan for it, which is why the hedgehog doesn't move, and it's very static, and it looks a little bit off.
And so, I don't want the hedgehog to appear here. The reason the hedgehog appeared here is because it's still connected up to the reference sheet. We don't mention it in the prompt, but because it's connected to the reference sheet, it's trying to use that information in the image generation. So I could simply click um delete on the connector and power of flows is now I simply click run from here and it regenerates everything after where I've clicked it. And so now it's redoing that image. It's redoing the video and it will then also recompile all the clips so I get the finished result in one single click which saves a huge amount of time.
Because before I would have to go back into image and video, re-upload my character reference sheet, change the prompt, make oh actually don't include the character reference sheet, but then turn that into a video, copy paste my prompt here. This is all done automatically and it saves a huge amount of time. And while this is regenerating, we want to also add a little uh voice over and we also want to add some background music. And once again, you can find um all of the tools inside of 11 Creative inside of Flows, but I think it would be fun if we actually went back to um 11 Labs just so I could show you how to create a really nice and compelling voice over.
So to create our voice over for our video, we're going to be using the text to speech tool. Inside of the text to speech tool, you can type out text. You can type out the voice over that you would like. So here I've actually just got a little voice over that we use to announce seed dance 2.0 that's available inside of 11 creative. And just to show you a quick example, let me just remove this. I've typed out the voice over. Um I've got a voice selected and we'll look at how to change it in just a second.
But if we click generate and let me know in the chat if you can hear this by the way. Hey everyone, just to let you know that seance 2.0 is available in 11 creative. As you can see, we've got a lovely voice over that matches exactly what we're saying. So, for my little hedgehog um film, for my short film, my trailer, I'm just looking for my prompt off the screen. I've written um this little script and let me just find it here. Perfect. Okay, so the story Little Wer. Excellent. Little Wer. Perfect. We can all hear it in the chat.
I think there's a little bit of delay, but thank you so much for uh for letting me know. Um, we've got this. This is the voice over that I would like to use for my Hedgehog short film. In a quiet corner of the world, a tiny adventurer takes her first step. One step, then another until the whole sky opens up. And what I can do first of all is change uh the voice. Actually, the first thing we're going to do is change the model because there's a couple of different models, but the model I recommend using to get the best and most realistic text to speech is uh 11v3.
And so, you can simply click Whoops. You can simply click 11v3 right here. And the reason we want to use 11v3 is that it's very expressive and you can use what you call audio tags. Audio tags allow you to guide and direct the delivery of your voice over to perfection. And so let's actually just quickly go and change the voice. And so I like um if I go into my voices, uh a voice that I'm a really big fan of is um a narrator voice. I believe I might already have it saved. Yes, it's David American narrator voice.
And so if I click on this voice, we can actually preview how the voices sound. Once upon a time in a quaint little village nestled among rolling hills and perfect voice to narrate my little hedgehog story. So I'd click on this voice and now we can add audio tags. So we can uh simply generate this. So just to to get a preview without audio tags. In a quiet corner of the world, a tiny adventurer takes her first step. As you can see, that's already an incredible voice over for um our story. But audio tags are little words and objectives that we can type in between square brackets to direct and guide the voice, just like a a director would tell an actor how to deliver a line.
So, all I need to do is open up a square bracket and let's say in a quiet corner of the world, we wanted this to be whispered. Well, I could simply type out whisper. And now, as you can see, it's been highlighted in purple. And so, the first sentence should be delivered with a whisper. And I can go and add in multiple different um audio tags throughout. And so, I could add one which gives us a little bit of a pause. I could add uh another one where we say gentle and you can kind of use any word to describe uh the delivery.
Um and so you can be really really creative here and get some great great great great results. And so for example, little wonderer, I might want this to again sound um affectionate. You know, we're telling the name of the story. But now that I've added these audio tags, the delivery should sound a little bit different and it should match the audio tags that we've added. So if I click generate. One step, then another until the whole sky opens up. Little wanderer. Is that not perfect? That sounds like the best voice over for my short story um about my little uh hedgehog.
And so I could I could then what I could do here is obviously I could go and add this uh to my files. I could download this. But again if we just go back to the flows inside of 11 creative um sorry inside of flows I just want to mention that you can um use uh text to speech directly in there. And so if I actually go and add a where is it text to speech node as you can see we can type in our prompts directly here. We can change the models. Um we can go and uh choose the voices.
So once again I could go and find the same narrator voice that I really like. I could paste in my prompt and then I can um connect this up. So we've got audio tags in our composition node. um audio tags in sorry we've got audio tracks a lot of words audio tracks in our composition node and I can click drag drop and now when I click run I can actually do this little arrow and I can do run from here so it will generate the voice over and then it will regenerate my composition node with the voice over and we could then go and do the exact same thing for music and while this generates let's actually go to Music.
Music. Another tool inside of 11 creative. We can click on music on the left hand side in the toolbar. And now we're taken into music. And there are first of all an incredible amount of generations with inside the music marketplace which allows you to go and um create generate and earn music and then remix other people's music. And it's just an incredible tool. You can create some incredible music. And so for the um for the way it works is exactly like image and video. We describe what we want to hear. Instead of describing what we want to see, we describe what we want to hear.
So at the bottom, we've got this little prompt box right here. So I would simply describe the track that I would like. And I've got a pre-saved prompt uh that I like right here. So let me just copy and paste that. Um and as you can see, I'm describing the use case that I want. uh this music for. So it's like for a warm whimsical story book score, right? I'm describing the type of song. I'm including um instruments, specific uh styles of instruments that I want to hear, sounds that I would like to hear. And a great thing to get some really good results with 11 music is to actually include the emotion that you would like to convey.
So, if you were looking for something um that was very happy or something very sad, actually including that in the prompt um can give you some really good um results. And then at the bottom um there's a bunch of different settings once again. So, we can choose the number of variations that we generate. We can choose uh the duration of our track. And so, mine just a short story, but I want to be safe. So, I'm going to generate one minute long. Um and then here, we don't want any lyrics. So, instead of leaving this on auto, because 11 Labs AI will basically detect whether it thinks you need lyrics based on your prompt, but I definitely don't want any lyrics.
So, I'm just going to change this to instrumental. Um, and then we're not going to use a fine tune. Um, and I won't go in depth into finetunes. You can find uh in-depth tutorials about fine tunes on our YouTube channel. And then we don't want to rewrite my music prompt because I'm pretty happy with the way it is. And so, here I would simply then click generate. And now we are generating the music for our short story based off our prompt. And this should usually only take a couple of seconds. Excellent. And now we've generated it.
And if I click play, as you can see, this sounds uh this sounds perfect for my short story. And so I could then go and download this. I could also um once again add it to my files and then we could bring it back into um flows. And once again, we could also do this directly inside of flows. I could create a little 11 music node or a music node. Um I could toggle lyrics off and I can simply describe the music that I want to generate and then we would run this and then I could connect it up with um a actually a new audio track.
So here I can click new audio track and this time drag in the music and we can adjust the volume. This is going to be background music. And so I would love to turn this down probably to about 16% based on my previous experience. But now if I toggle this off, we won't have the music yet, but we should have the voice over, I believe. Um, let's wait for it to restart. Here we go. adventurer takes her first step. One step, then another until the whole sky opens up. Little wanderer. Perfect. So, we've got this great uh voice over.
It's not 100% in sync with um the the animation because here we're in the composition node. If we want to take this one step further and go uh like in depth on the editing, we could then move over to studio. And so I like to think about it this way. There are two different interfaces that you can use inside of 11 creative that allow you to use all of the 11 Labs uh or 11 11 creative tools. So if I just quickly go back um to the dashboard, right, all of these tools down here are individual tools that you can use inside of 11 creative.
like we've mentioned, text to speech, 11 music, image and video. And then up here we've got the two interfaces. Flows, which is the infinite canvas that we've just been using. And then we've also got studio, which is um a more traditional editor, but still built with AI first in mind. And again, it has access to all the tools that we've got access to individually here. And I realize I haven't been answering that many questions, but yeah, it's Brian has just asked if lip sync is possible. And yes, lip-sync is 100% possible. again available in all of the tools and funnily enough um there is a lip sync tutorial that you can find on the dashboard and also on the YouTube channel um right there so yeah lip sync is definitely possible but everything I've shown you up until now has been inside of flows if we had wanted to have more control over the structure and the composition and maybe we wanted to splice up our voice over we could go into studio and so here I can go into studio and I could simply click uh new blank project and we You can actually do this uh live.
And so obviously I don't want to recreate everything, but I do want to reuse the assets that I've generated inside of my flow. And so what I can do is I can right click on my little hedgehog and we can do save to files. And I'm just going to save this into hedgehog story. And literally just click save like that. And I'm going to quickly click save on each of these. Only takes a second. So I don't have to, you know, download them, re-upload them, and I can keep everything very, very organized. Um, and so I'm just doing save.
And then last one, save to files. Hedgehog story. Save. So now in my files, as you can see, if I open up Hedgehog Story, we've got all of the generations that were made inside of Flows. And I can then reuse these inside of Studio. And so, let me just quickly uh refresh this. Um um what I'm going to do is I can't seem to Yes, I do have them somewhere. I have I had files at some point. Um but they don't seem to be coming up. Yes, they are here. I'm just being silly. Sorry. Files.
Um it's been a long day. So files. If I click on Hedgehog story, I have all of my generations. And so I can simply go through each one, click on the little plus sign to add it to the timeline. And now I'm adding all of my little clips. And so I'm just going to quickly put them one after the other. And so this is very much like your traditional um video editor and timeline. And so I can click and drag and move these all around all around and compile them just like um a classic video editor.
And so you might have done an AI generation that you love the first 3 seconds of or you you think it it you love to be able to trim off the beginning because it doesn't quite work or it doesn't quite you know match the length of your music. You don't necessarily want 5-second scenes every single time. And so you can and let me just finish uh structuring these. But let's say for I'm just putting this over here. Let's say for the before last scene. So, this one right here, we didn't like the fact that it lasted so long and I didn't want this the first few seconds.
Basically, I could simply trim um by clicking and dragging just like so. And now, instead of it being 5 seconds, it's about 4 seconds. And then I can just quickly rearrange it and then bring on the the final part right there. And I had a little gap. So, I'm just going to close that. But what's really cool um is that obviously we've got uh the voice over that I want to use and also the background music. And so again on the left we've got all of the tools that are available individually inside of 11 creative inside of studio.
And so let's say we went to uh where is it? Let's go to speech. Um here once again I can go and generate the uh voice over that I like. And so let me just quickly grab a prompt. Actually, I'm just uh Yeah. All right. Cool. Let me go and, you know, regenerate something a little new. Um, cool. No, I don't want that to be individual. Whoops. One second. Actually, I know what I'm going to do. I'm going to quickly go back to home. We're going to go back to text to speech. Excellent. And I'm going to reuse this exact prompt or just to show you a different way to do it.
Um, if I generate it, because obviously we can add everything to files in a quiet corner of say I really like this one. Um, once it finishes generating the entire thing, I could then actually download it. And so now this is downloaded to my laptop. Um, and I could also drag in this audio file into flows. And so actually, let me just go to files. Here we can click upload. And you can't see my little finder, but here I'm just selecting the download that we've just done. And this is great because sometimes you have your own assets from somewhere else.
You might have got have your own voice over or you might have if you're making ads, for example, because again, we can apply this to all different types of videos. If you're making ads, you'll have your own brand assets, your own logos. You can upload all of your own stuff. And so now I've got this um this perfect voice over. And I'm actually just going to delete uh all of this right here. Now I've got this voice over. Um, and again if we play it in a quiet corner of the world as you can see um, we've got our video, we've got our voiceover.
Last thing I want to do now is just quickly add music and I aware we're coming up to 46 minutes. So I appreciate people being very patient and going through this. And we are coming close to the end. So if you do have any questions, please do get those ready. Um, but now if I click on music, as you can see, we have actually got the history of all of the generations I've done. So this was the one we actually made earlier. Perfect. Oh, there's this one as well. Or we've got this one though. I quite like this one.
It's a little bit more uptempo. Um, fast-paced. I like it. So, again, I could just click the little plus plus icon. And boom. Now we have our background music. And obviously, it's going to be too long. So, I'm just going to trim this just like so. And then what I am going to do is I want to add a little fade out because when we get to little wonder, we say little wonderer, I want the music to slowly fade out. And so up here I can simply uh make a fade out. And as you can see as I click and drag it, it affects the sound waves just to let me know that it is being applied.
So let's just say I could add it for let's just do about 5 seconds. Now, the issue we were having earlier with the composition node is that we wanted to be able to, you know, trim things, adjust, and I wanted the voice over to um to have longer pauses. So, when we got to the end title screen, Little Wer, um it matched because right now, Little Wanderer, it doesn't match, right? So, what we can do is I just place the play head where I would like to make the cut of uh the voice over and I can click this little button right here, and I simply click it.
And now we've actually split the audio file. So I can then move this and drag it and time the voice over to perfection. So if I click play now, little wanderer, that's perfect. You know what I might do is quickly increase the fade out and then lower the volume because this is background music. I don't want it to be too present throughout the video. And so I'm going to put this down to 10%. And now, ladies and gentlemen, if I play this from the beginning, um, actually, you know what? Being a perfectionist, I might just add a little pause in here and make this a little wider.
And then little pause in here. And I'm not even listening back to this, but I think it will make it more engaging. Excellent. Now, let's watch back the AI short film that we have just made um, called Little Wonder. Little wanderer, how cool is that? That's really cool. That's really, really cool. And so you can then go and take this much further and create much longer AI videos using consistent characters through having your own reference sheet and using flows to be able to create it much much quicker. And so yeah, exactly. We could create um what we said.
Yes, it's Brian. Let's dream big. We could create a full TV show. Yeah, you can create full TV shows. And there are people that are using 11 Creative to make incredible uh incredible pieces of video, incredible short films. Um, and so that's it. That is all you need to know to kind of jump in, get started, and start creating your own content using uh 11 creative. And again, everything that I've taught you for the text to speech, uh, the prompting, the model selecting, um, the music generation flows, studio, all of this can be applied to various different types of video.
So, we could quickly imagine how we could start creating um ads at scale with multiple uh different voiceovers and and and generate those. We could start thinking about how we could create um the voiceovers for YouTube videos. And uh I've been a 11 Labs user for three years. I've used voiceovers a lot in my YouTube content. Um and so there we go. I think we're wrapping up to uh the end, but if you have any questions, now is the time to drop them in and I will see if I can ask them. Otherwise, don't hesitate to ask us on X on the 11 creative account.
And speaking of the 11 creative account, to celebrate, have I disappeared off screen? I think I'm coming back. Excellent. to celebrate uh speaking of the 11 creative account to celebrate the release of GPT image 2 and the fact that it's now available inside 11 creative. If you go to 11 Creative um you can actually go and access um or you can get free credits, 5,000 free credits to try out the model to try and create some AI video, some voice generations. So, uh yeah, check that out, find it on X. But otherwise, thank you so much uh for joining.
Um, thank you, LSU. Thank you, Sync Lexia. And, uh, yeah, it's Brian asked, "Will this be on replay or on demand?" You know what? I don't think so, but everything we've covered, we we might be able to figure it out with the team. I'm not too sure. I think it is being recorded somehow. Um, but everything that we've covered, you can find on the 11 Labs YouTube channel. And, uh, we actually I' I've just got word from Matcha. So, thank you, Matcha. A lot of you will know Matcha from from X. Um, he's saying that it will likely be on YouTube for rewatch.
So, I hope it was helpful. And if I didn't answer any questions live, ask them in the comments on that video. But in the meantime, um, head to uh head to 11 creative on X and claim your 5,000 free credits. Thank you very much everybody. I think that concludes the live stream. I hope you all have a wonderful rest of your day and uh I will see you on the next one.
More from ElevenLabs
Get daily recaps from
ElevenLabs
AI-powered summaries delivered to your inbox. Save hours every week while staying fully informed.




