FREE AI Voice Tool: Text-to-Speech (TTS) & Voice Cloning - MetaVoice

this is absolutely insane as I have just found one of the best AI for human level speech conversion tools introducing metav voice in other words it's text to speech model that is completely for free with really great AI voice generation just take a look at this demo after graduating from Cambridge with thoughts of his father hanging over him Hinton moved to London and became a carpenter it wasn't fancy carpentry he says it was carpentry to make a living that here he read the organization of behavior a book written by a Canadian psychologist named Donald H Now isn't that amazing metav voice 1B is a 1.2 billion base model that's trained on 100K hours of speech for text of speech it has been built with four key priorities firstly you have emotional speech Rhythm and tone in English with no hallucination so this basically means that whenever you're cloning a voice with their AI voice cloning system you're going to have zero hallucination in the generation secondly you have zero shot cloning for American and British voice with 30 seconds reference audio sorry for being repetitive but this month we had insane Partnerships with big companies giving out subscriptions to AI tools completely for free these are tools that will streamline your business's growth and improve your efficiency just being a patreon this past month you were given access to six paid subscriptions completely for free not only do you access these subscriptions but you gain the ability for Consulting networking collaborating with the community as well as with myself you get access to daily AI news resources giveaways and so much more if you're interested check out the patreon link in the description below to gain access to these benefits audio thirdly you have support for cross lingual voice cloning with fine-tuning which lets you fine-tune different types of accents with different cloning methods within metav voice fourthly you have the priority to be able ble to support for long form synthesis this is all done with metav Voice's model now after using it myself with the Google Cloud as well as with their demo I was able to actually see how scary this could actually become this is a model that's under the Apache 2.0 license which basically means that you can use it without any sort of restriction and it's completely for free now throughout today's video we're going to be exploring mavo 1B further in detail by exploring its capability showcase how you can get started and just go over some demos so with that thought guys stay tuned and let's get straight into the video if you would like to book a one-on-one with me where you can access my Consulting Services where I can help you grow your business or basically give you a lot of different types of solutions with AI definitely take a look at the calendar Link in the description below hey what is up guys welcome back to another YouTube video at the world of AI and today's video we're going to be taking a look at a new AI voice cloning model called metav voice now before we even get into the architecture or the capabilities I just want to talk a little bit more about what sets metav voice apart from platforms like 11 Labs or tortoise now TTS is quite a unique combination of features with all of these platforms but what sets metav voice apart from all of those other platforms is the training data and the model size of metav voice because it's something that boosts a 1 Point 2 billion base model trained on 100K hours of speech data this is quite extensive and it basically minimizes hallucination and it doesn't require you to basically input a lot of samples as they have a zero shot cloning ability which lets you only like input around 30 seconds of reference audio to get a good cloned voice out of it so you might be wondering how can you actually get started well it's fairly easy you can deploy it on Google Cloud which is possibly one of the easiest ways to get started this is where I'm going to be showcasing this as we go further into the video second method is trying the demo out to get a better feel as to what you can do with metav voice and lastly you can have it installed locally they have a good guide as to how you can do this so if you're interested take a look at the reference implementation plan which showcases how you can install this locally they also have a instruction manual as to how you can deploy it on cloud with AWS gcp as well as aour so I'll leave this link as well as all the links that I use in today's video in the description below now how can you actually deploy this on Google collab well it's fairly easy what you want to do is click on file and save a copy in your drive once that is done go over to runtime and change the runtime type the best hardware that is available once you have that set you can then go forward and install each tab that which will be required to have this functional now if you go onto this Google collab you can see that they have basically showcase a couple examples where over here there is something that has been able to generate all these voice like files and we can see it's based off of this one single prompt over here which is stating clone this prompt which states that high Sam this is a demo of text to speech by metav voice 1B and open source foundational Audio model by metav voice and if we are to go down we can see that it generates a quite good Voice demo out out of it Hi Sam this is a demo of text to speech by Metabo 1B and I believe it does it in various examples Hi Sam this is a demo of text dis speech by metav voice 1B and we can take a look at this one Hi Sam this is a demo of text to speech by mat voice 1B and lastly let's take a look at this one same this is a demo of text to speech by met voice one b and it seems like there's different types of Stu Styles as to how they pronounce the words different gender speaking and so much more so you can see that there's a lot of customizable ways to have this working now this is a notebook that was actually created by a YouTuber called Sam and I'll leave his video showcasing this GitHub repo or sorry not repo but this Google collab notebook cuz he goes more in depth that's how you can use it but this is a good way to get started with this cuz he has uploaded the code for metav voice on Google collab so let's see clone voices straight from metav voice or sorry from Google collab now if we go down even more there's more examples and I believe he clones it in different ways with different sorts of samples where he uploads 30 seconds of a sample and has a cloned cloned referencing that sample now after you have installed all the blocks what you need to do next is set your output directory and this is by clicking on this button over here making sure that you you have your output directed to the same folder so once that is set you want to then upload your samples this is where you can upload samples as to how you want your voice to be cloned or what text you want it to be cloned off of and you can basically then upload the sample data have it so that the sample is connected to this block and you can do that by simply just replacing the location or where the file is coming from from this file tab over here and you can simply upload your files by clicking on this button and putting it into here now you just require approximately 30 seconds of an audio file and you're going to be able to generate it and you can see that after you input the text that you want to basically clone you can then have this block runed and then you can set this other block running and then you will get this output over here which will then be saved in your samples and or I believe it's this in one of these two tabs but it'll be outputed in these two tabs and then you'll be able to download it and use it whenever you want now before you even get to Google collab I recommend that you play around with this on the demo uh which is something that they have uploaded and it's currently free to use and this is where you can input a prompt so in this case I want this to be generated in terms of an AI voice and this is where I stated that my name is world of AI and I love making YouTube videos on AI I set the parameters and I kept it the same basically and then I chose to just present the voice in any sort of natural way that you want but in this case if you want to upload your own voice where you can have a sample based off of that voice generated then you can set it over here by clicking on this button but in this case I want to be presented with Bria but you can also select Alex or Jacob and then once that is done you can then click generate speech and let's see how this sounds I'm just going to turn this down because it's going to be kind of loud my name is world of AI and I love making YouTube videos on I now that was kind of a little too fast but I can adjust this and I can have it so that it doesn't speak it super fast now let me see if I can like lower this and let's see how it sounds my name is world of AI and okay no that doesn't sound good but you get the gist of it it's able to generate and it's quite humanik now in this case it was kind of too fast but we can adjust it so that it's not too fast and it'll be able to sound pretty human-like and I love making YouTube videos on I and that's about it for today's video on metav voice guys I hope you enjoyed it and you got some sort of value out of it this is quite an amazing new AI voice model and I truly recommend that you check it out with all the links that I used in today's video but with that thought guys thank you guys so much for watching I hope you enjoyed it make sure you check out the patreon page if you want to access our private Discord make sure you follow us on Twitter this is a great way for you to stay up to date with the latest AI news and lastly make sure you guys subscribe turn on the notification Bell like this video and check out our previous videos so you can stay up to date with the latest AI news but with that thought guys thank you guys so much for watching have an amazing day spread positivity and I'll see you guys further shortly peace out f

Channel: WorldofAI

Views: 7,725

Keywords: ai voice cloning, ai voice, ai tools, voice cloning, voice cloning ai, free voice cloning, ai, realistic voice cloning, clone your voice, voice cloning tech, text to speech, tts, meta voice, meta voice 1b, voice, bark, text-to-speech, speech-to-speech, speech to speech, voices, ai voice generator, best text to speech software, free ai voice tool, bark ai, musiclm, music, speech generation, amphion, clone your voice ai, clone your voice for free, clone your voice and make it sing

Id: gVKbf31hrYs

Length: 10min 4sec (604 seconds)

Published: Fri Feb 09 2024