🚨 RECRIARAM A ELEVENLABS! Agora Você Pode Usar de GRAÇA 😱

Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.

Original language: English

Original video: 🚨 RECRIARAM A ELEVENLABS! Agora Você Pode Usar de GRAÇA 😱 · Channel: You Ensina · 10:56 · Watch on YouTube

Topics: AI content generation, Local and open-source models, Multimodal models, Open source and software ecosystems

Summary

The video introduces Voice Studio, which the presenter describes as a free, open-source, local alternative to ElevenLabs, and demonstrates how to download and install it. The walkthrough covers setup options, voice generation, cloning, and other features such as dubbing and transcription. The presenter tests voice generation but reports that the Portuguese output sounds garbled, suggesting that language settings or voice details may affect the results. The video concludes by inviting viewers to try the tool and share their experiences.

Key points

The presenter says Voice Studio is a free, open-source alternative to ElevenLabs that runs locally.
The video directs viewers to download the application from its GitHub releases, with versions for Windows, Mac, and Linux.
The setup process installs a local environment and dependencies, then offers model options with different resource requirements.
The presenter chooses the fast model for the demonstration and notes that other models offer additional features.
Voice Studio includes tools for voice generation, voice cloning and design, dubbing, transcription, stories, and audiobooks.
Users can adjust voice settings and configure details such as a voice’s name, transcript, style, and language.
The presenter’s Portuguese voice-generation test produces garbled audio, and a second attempt also sounds unusual.
The presenter suggests setting the voice and its details to Portuguese, while acknowledging that the cause of the poor results is uncertain.

Chapters

Start	Chapter	Summary
00:00	Introducing Voice Studio	The presenter introduces Voice Studio as a free, local alternative to ElevenLabs and says the video will demonstrate how to install and use it.
02:05	Installing and configuring the app	The walkthrough covers language selection, local setup, environment checks, and choosing a model with suitable resource requirements.
04:41	Exploring voice tools and generating audio	The presenter explores voice settings and the app’s features, then demonstrates voice generation and cloning.
08:21	Testing results and closing advice	After the generated Portuguese audio sounds garbled, the presenter discusses possible language-setting issues and wraps up by inviting viewers to try the tool.

Quotes

They named this tool Voice Studio; it’s an open-source, fully local alternative to ElevenLabs. — 00:00

Here we have versions for Windows, which is the 64-bit one, versions for Mac, and also a version for Linux — 01:33

Now I can say whatever you want, using your voice or mine. — 06:16

the voice came out all scrambled — 08:21

put everything here in Portuguese so the tool can create it the right way for you. — 10:25

Concepts and entities

Voice Studio, ElevenLabs, GitHub, clonagem de voz, dublagem, modelos locais

Full transcript (original language)

00:00 That’s right. They reverse-engineered ElevenLabs and released it for free. Come see how to use it. That’s right, my friends. It’s bad news for ElevenLabs because someone reverse-engineered it and released it for free for us on GitHub. And we have to take advantage of this, guys. I translated it into Portuguese here for you to take a look. They named this tool Voice Studio; it’s an open-source, fully local alternative to ElevenLabs. As you can see, the creator himself

00:31 says here that he reverse-engineered ElevenLabs. Voice cloning, voice design , video dubbing, dictation, transcription, and audiobook creation in 646 languages. All of this for free for us, as he mentioned right here. Completely local for us to enjoy. And we really have to take advantage of this, folks. I’m already providing the link for you in the description below, but stay in this video to learn how to download, install, and create everything from scratch. Got it? Here

01:03 are all the files. As you can see here, this was the person who did the reverse engineering. But downloading is quite simple. The first thing we need to do is install Voice Studio, which is the new name they gave it. So, we need to scroll all the way down, guys. There is a lot of information he leaves for us, but we need to go down to the bottom and click on this option to download. Download from the releases section. You click right here. Excuse my English, as my English is terrible. Then it will

01:33 redirect us here. Here we have versions for Windows, which is the 64-bit one, versions for Mac, and also a version for Linux for you to download, all of this once again. For free. After you download it, it will come here. As you can see, you perform the installation normally on your machine. Got it? Well, after you download and install it, right? It will open up here on your machine. Here it gives you the language option. It covers 646 here, if I’m not mistaken, as I told you before.

02:05 Languages. So, here, look, we have Portuguese, folks, it’s better for us . And the region here, look, we can just leave it on automatic. It gives some options here, but just leave it on automatic. Got it? Here we can click, look, install local environment. Here it will check the environment and download the file manager. It will start installing several files right onto your machine. It might take a while, you know, to finish this installation, because there are many files, like I’m telling you, since it’s open source, right? So it goes

02:36 directly into your machine, your PC, and it needs to download many files. So please be patient while it downloads everything, deal? Well, my friends, it’s finishing up here, look, all the installations. It downloaded 217 dependencies and was installing them, as you saw, now it’s running an environment check here, right? And it does all of this automatically. Well, after that part, it comes here, right, to do the checks. You can see my GPU here, look, it gave a small warning because mine isn’t one of the best,

03:07 right? But even so, I’m still doing it to show you the potential you can use. Otherwise, everything else is fine , right? We can, look, okay, continue here. It adjusts for Windows 64, right? Check the local models for disk usage, installation, and activate the set of steps. Now here we have the fast model to use, the balanced model, which is already 4.8 GB here, look, and quality which goes up to 8. And the maximum, which goes up to 8.5 as well. Let’s put

03:39 it on fast, because that’s what will consume the least for us, right? we click here to install as well. It will perform the installations here. Remembering that if you want something more, right? There is the balanced one here too that you can use, it adds this dictation option too, this one, right? In quality, it also covers the translation option. So, if you want something more, as I told you, it will consume more gigabytes on your PC. And so, I’ll do it in fast mode here. Maybe later I’ll even add the balanced mode . I was supposed to have done it in

04:09 that one. I ended up doing it in fast mode on impulse. But I’ll do this one here. Show you how the result turned out. So we can continue by clicking here. All right? Continue. Here it asks , see, if we want to help improve Voice Studio. Basically, it’s asking us, you know, to share information? Like, yes, share anonymous statistics; you’re not obligated to share, and you can click no here. Right? Right here. All right? Continue. And here is to enter the studio, right? We can click either here or here, see. Enter studio. It will come to this part here, but you can

04:41 clear it up here at the top, see. And there you go, you’ll be inside the tool . And here we can start using it. Well, let’s start creating something here. Here it asks us, see, to describe your voice. We can do the creation, right, see. Describe the voice or choose a starting point. Here we have the profiles, narrations, right? Look at how much information we have on our screen, guys. Here on the side we have the languages, so we can set it to Portuguese here. Here we open these settings, right, such as speed,

05:13 stability, temperature, layers, right, automatic duration, which we can configure however we want. Well, it’s generating an audio here, as you can see. Now, we just have to wait, right? While it’s generating here, I’m showing you that the tool is providing here on the top sidebar, see, it has these options for stories, audiobooks, right? You can also see it features dubbing, right? Since it has the option to dub and batch dub. Here is the option to transcribe. I even changed pages here,

05:45 I’ll go back there, right, here is the transcribe option, there are the projects, right? Tools and integrations . And up here, see, it has the option for voice cloning, design, save, and gallery. You can see that it covers a lot of things here that ElevenLabs itself has brought here for us for free , right? Well, let’s open this so we can see it properly, right? All these options. Look at all the options here, right? Here we have the history down below, see, right? Well, and take a look at how our audio creation turned out, but we are going to create one together. Take a look at how the result

06:16 turned out. >> Welcome. A moment ago, I was just a 3- second clip. Now I can say whatever you want, using your voice or mine. As you can see, right? Here, there is an option to save how cool the voice sounds, right? Oh, remembering that we only downloaded the quick model, right? So, we are making these types of creations here. Well, I will click save here. We can save it wherever we want. I’ll just save it in downloads. Let’s create one from scratch. I put a little sentence here, saying I am 25 years old and a little thank you. Down here,

06:48 before we create it, we have the languages, as I showed you, right? There is also the settings section that you can use, which is important. Creating is very simple. Click here, and it will start the creation for you. Another cool thing is that if we come here, to voice sample, and click on it, the voice that was cloned inside the tool will appear. You can see that we are in the cloning section, right? So, this voice was cloned, and it made it available for us to run these tests. But you can upload the voice of whoever

07:19 you want here. For example, many people use Valentino’s voice, Adam’s voice, right? You can also upload voices of famous people if you want, right? Remembering that this is not a recommendation, guys. Another very important thing, you can click here, edit voice, and here you add the voice name, add a transcript, add a style, language, right? Here, below, there is export persona, test voice, voice properties. So you have full configuration control over the voice you want to clone and use. Got it? This

07:50 is very important so you can get a good and very realistic voice, right? To avoid those robotic voices. Let’s go back here to generate our voice, right? Let’s wait for it to finish the creation. The other cool thing I forgot to show you is that down here, there are several integrations they make available for us to use. Many integrations indeed, like cloud code, right? There is even Voice Studio here, API, API format that you can also use. So these are also options that, if you are interested, you can do. Well, guys, after a few seconds, it delivered this

08:21 result here. Let’s see how the generation turned out. Remembering that this is the same voice the tool gave us itself, right? Let’s see how it turned out. A bot that sounds like >> Well, guys, as you saw there, the voice came out all scrambled, right? So, it wasn’t a good result like we expected. I set it to generate again to see if it can deliver a good result, because this one really didn’t turn out well. The speech was kind of scrambled. I don’t know if it didn’t understand what I put here or if I have to put it in another

08:51 language, or if the voice itself is from another language. And since we put Portuguese here, it didn’t manage to detect what we asked for correctly. So there are those considerations. Make things with more detail here so it can better interpret your creation, because as you saw here, it has potential, right? It has great creative potential, which, as I told you, is like a cracked version of Eleven Labs. The guys reverse-engineered it and made it available for us here. Of course, there will be bugs, just like Eleven Labs

09:23 itself has bugs, right? So, like, these kinds of things can happen. Comment below if you already knew about this tool, this reverse-engineered version of Eleven Labs. Comment if you already knew it and if it worked normally when you used it. Well, I’ll wait for it to finish this generation and I’ll come back to show you how the result of this new generation turned out. Well, guys, no sooner had I finished talking to you than it finished right here, look. It really just finished. The little message that was here has already disappeared. Let’s see now if it turned out good or not, right? This one was

09:53 more spaced out. Voices and weird sounds. >> Yeah, guys, it didn’t turn out as good as I expected. I think it might be the language here, right? Because this is a voice cloned with artificial intelligence that they left here, right ? I don’t know how the first one I showed you turned out good, since it was also in Portuguese, but now it didn’t, did it? You can see, look, that it’s kind of weird, right? It’s really weird. So when you go to do your voice clones, put a Portuguese voice here, in

10:25 the Portuguese language, right? Put the edit here, add a few things, see, the transcript here is in English, that might be why, right? We don’t know. Got it? But put everything here in Portuguese so the tool can create it the right way for you. The concept of this video is to show you this tool called Voice Studio, which has been reverse-engineered so I can show you that it can handle these creations for us, right? There are many options here, just like I showed you, for stories, right? Many options that used to be

10:56 from Eleven Labs and are now available for free to us. Got it? Well, comment below what you thought. But here, basically, I’m thanking you all for the 20,000 subscribers here on the channel. Thank you so much to those of you who are supporting my work and who are enjoying my content, right? So, thank you very much once again. We’re in this together. I’ll be heading out now. Comment below if you’re going to test this tool, got it? And see you in the next video. Remembering that its link is in the description over on GitHub. Got it? See you in the next video.