How to Use ElevenLabs AI (Complete Beginner Tutorial)

[00:00] Creating professional voiceovers no longer  requires expensive microphones or a recording   studio. That voice wasn't recorded by a person. It  was generated entirely with AI using ElevenLabs.  

[00:13] I'm Elizabeth and today I'll show you how to use  ElevenLabs to create realistic AI voiceovers,   clone your own voice, dub videos into other  languages, and build long narrations. The first  

[00:27] thing you'll want to do is head to ElevenLabs.io  and create an account. The link to sign up is in   the video description and pin comment below. If  you're just getting started, the free plan is a  

[00:39] great place to begin, and it's what we'll be using  today. It gives you enough usage to experiment   with many of the platform's most popular features  before deciding whether you need one of the paid  

[00:51] plans. Once you sign in, you'll arrive at the  dashboard. Now, if this seems overwhelming,   don't worry because in a few minutes, everything  here will make sense. Along the left-hand side,   you'll see the different tools available.  Depending on when you're watching this video,  

[01:07] you may see slightly different options since  ElevenLabs is constantly adding new features,   but the core workflow stays the same. We'll  get started today by selecting text to speech.  

[01:19] Everything here is designed around a simple  idea. You type text into the box, choose a voice,   then ElevenLabs generates an incredibly  realistic narration. So, let's see it in action.  

[01:33] I'll start by pasting in a short script. But  note that you can include up to 5,000 characters.   And that's all I need to do before I generate  my speech. Nestled along the rugged coastline,  

[01:46] the small fishing village has welcomed travelers  for generations. Colorful boats sway gently.   That's pretty impressive, but probably not  the right tone of voice for this narration.  

[01:58] So, let's dig into the voices in more detail. One  of the things that immediately impressed me about   ElevenLabs is just how many different voices it  offers. And you can browse voices based off of  

[02:11] language, accent, gender, or even the type of  content that you're creating. And each voice   offers a preview so you can hear it in more  detail. I I totally understand how you feel.  

[02:26] Would it work if you tried talking with them  again? The clock ticked steadily, marking the time   until a revelation that would change everything.  You can also search for specific types of voices.  

[02:40] And a fine maro to you there, Captain Marshall.  is I finding a voice that matches your project   usually only takes a few minutes. After narrowing  down my options, I've selected Ricky Johnson,  

[02:55] an older man with a southern accent. Nestled along  the rugged coastline, the small fishing village   has welcomed travelers for generations. Colorful.  If you've used older text to speech tools before,  

[03:08] you probably noticed that they often sounded  robotic or unnatural. ElevenLabs does an excellent   job adding realistic pacing and emotion, making  it much harder to distinguish from a real person.  

[03:21] But let's make this sound even better. Below the  voice section, you'll notice several settings that   control how the AI speaks. The first is speed.  Exactly what it sounds like. Move the cursor  

[03:34] to the left and the speech will get slower. Move  it to the right and the narration will speed up.   Next, you'll see options like stability,  similarity, style exaggeration, and depending  

[03:47] on the AI model you're using, expressiveness.  These control how consistently the AI delivers   your script. For example, higher stability creates  a more predictable voice. Lower stability allows  

[04:03] for more variation and emotion. There's no single  best setting here. If you're narrating a corporate   training video, you might want a steady,  consistent delivery. If you're telling a   story or recording an audio book like I am, adding  more variation often sounds much more natural.  

[04:23] I usually recommend experimenting with these  settings until you find a style that you like. Nestled along the rugged coastline, the small  fishing village has welcomed travelers for  

[04:35] generations. Colorful boats sway gently in  the harbor as locals prepare for another.   And one tip that makes an even bigger difference  than any of these settings is your writing.  

[04:47] AI voices perform best when your script sounds  like something a real person would actually say.   For example, use commas where you would naturally  pause. And don't be afraid to add punctuation  

[05:00] like question marks or exclamation points.  Sometimes simply changing the punctuation   can completely change how their narration sounds.  Now that we've created our first AI voice,  

[05:13] let's take things a step further with voice  cloning, which is the ability to create an   AI version of your own voice, and it's included in  the starter plan. To get started, click on voices.  

[05:27] I'm going to use instant voice clone, which is  the fastest way to create your own AI voice.   The process is surprisingly simple. All you  need to do is upload a short recording of  

[05:39] yourself speaking. Once you've uploaded  your recording, you'll name your voice,   agree that you have permission to use the  recording, and then click on save. After a few   moments, your new AI voice is ready to use. Now,  let's head back over to text to speech. This time,  

[05:57] instead of selecting one of the built-in  voices, I'll choose my newly created model,   I'm going to use the same script as before so  that we can compare the results. I'll select this  

[06:11] and then click on generate. Nestled along the  rugged coastline, the small fishing village has   welcomed travelers for generations. Colorful boats  sway. It's pretty remarkable how closely that  

[06:24] resembles my voice. Remember, no AI voice clone is  going to be perfect, but this one is pretty close.   And you can also make adjustments to your cloned  voice model similar to what we did before,  

[06:38] including the speed, stability, similarity,  as well as the style exaggeration.   Now that you know how to generate speech and even  clone your own voice, let's look at how to use  

[06:57] build an entire project in one place. You can  either upload a document like a script or you can   write or paste by clicking on new blank project.  Today we're going to do an audio only project.  

[07:14] I've pasted in a longer script. You'll notice that  ElevenLabs automatically separated the text into   different sections which makes editing a lot  easier. For example, if one line doesn't sound  

[07:28] quite right, you don't have to regenerate the  entire project, you can simply regenerate that   one section until you're happy with the result.  But my favorite feature of the studio space is  

[07:41] adding different voices for different characters  in my script. For example, I am starting off my   script with a narrator. So, here I'm going to  highlight which lines I want the narrator to say,  

[07:56] and I'll select the voice I want to use. You can  use your clone voice that you previously created,   explore the default voices that ElevenLabs  offers, or explore the library similar to how  

[08:10] we did before. I'll select Bella as a narrator.  And you'll see that Bella's icon has been added   next to those lines to help me keep track. Then  I'll be able to go through my script line by line  

[08:24] and pick the different voice character  for each one of those narrations.   And to help keep it easier for you, each one of  your used voices will be at the top of the screen.  

[08:37] The airport bustled with travelers heading in  every direction. Ed, do you think grandma made   her famous cinnamon rolls? Knowing grandma,  I'd say there's about a 99% chance. Only 99?  

[08:52] Well, there's always the chance Grandpa ate them  first. And once everything sounds the way that   you want, [snorts] you can simply export it  and ElevenLabs combines everything into one  

[09:04] finished audio file that's ready to use in  your video editor, presentation, podcast, or   wherever you need it. As you can see, ElevenLabs  offers a ton of different tools to play with.  

[09:19] that are available. And the first one of those  is Voice Changer. Instead of typing text, Voice   Changer lets you upload an existing recording  and transform it into another voice. Here's the  

[09:37] original audio recording. Good morning and thank  you for taking the time to meet with us today.   Today I'd like to show you how we're helping  organizations solve real business challenges.   And now here is the voice cloned that I just  created. Good morning and thank you for taking  

[09:53] the time to meet with us today. Today I'd like  to show you how we're helping organizations   solve real business challenges. Save valuable  time. I think this one is pretty incredible.   And of course you don't have to go with your  clone voice. You can explore all of the voices  

[10:09] in the ElevenLabs library. The next feature  is called dubbing, and this is great because   it can dramatically expand your audience. Let's  say you've created a YouTube video in English.  

[10:22] Instead of recording the entire video again in  another language, ElevenLabs can translate and   dub your video while preserving much  of the original pacing and emotion.  

[10:38] I actually think David sounds great speaking  in French. As ElevenLabs continues to evolve,   so it's worth checking back from time to  time to see what's new. Thanks for watching.  

[10:52] Let me know in the comments what you're using  ElevenLabs for. See you in the next video.