Local Voice Cloning Using XTTS Api Server

I spent hours on end studying so you don't have to learning which is the best text speech for your future projects listen to this I found out the food in your fridge or cooler decomposes I found out the food in your fridge or cooler decomposes also you stink a little anyone can install this program and if you have minimum of 4 GB of vram this will work on your PC this tool can easily clone any voice use it as a text to speech with less than minute of audio Source knowing this you don't have to go through tedious process like other backend programs having to train your own voice models also the real factor of its importance is how easy and fast it is to Output clone text to speech audio when combined with a chat gbt or a local llm in which I will be coming out soon with a video showing how later let's learn how to install this TTS this is a very important step you don't want to miss installing C++ Microsoft build tools is a crucial step to install this TTS API server once you managed to install it you now need to create a virtual environment so this is the command to create a virtual environment with python the other way to do it is with cond but we're going to keep it simple today and once you have enter this command you want to go ahead and activate it now this is the command to activate the virtual environment in Python once you have activated the PCH environment you want to go ahead and go to the next step to make it easy for you I'm leaving the commands in the description okay so the first thing you want to do after that is installing the TTS via pip and that is going to take a while so this is the next step okay so once you have the PIP installed for the API server you want to go ahead and install torch as well all right so to install torch you want to go ahead and go to the P torch website and install locally now this is very simple you just install for Windows if you have Windows Mac for Mac Linux for Linux and if you have CA 11 installed or CA 12 and then you just copy this down here now go ahead and enter that copied command into your activated environment and of course like I said earlier this will take a while so once you have everything installed you will have something like this so these two batch piles I will have it on my patreon but basically it's the same thing so in this environment there's going to be a command that you want to use so in this video we're doing TTS to file I will leave the command in the description all right so here in the fast API um web page we could use this our advantage to set our folders so we have right here post to set output folder and speaker folder so this is what these two do so now what should you want to do is go ahead and click on it try it out and in this string you want to go ahead and grab the folder you want to use for your output so for my instance here's my output with a bunch of audio already um you just want to go ahead and click up here and copy and bring it down here just paste it now this is not not it you want to go ahead and fix these slashes I'm I'm pretty sure and then execute now you will know if you did it right if you have successful response right here code 200 you could also check in the server right here 200 okay set output HTTP so that's one way to know that you did it successfully okay okay now you want to do the same exact thing for set speaker folder but with a different folder so for me I have my speaker folder right here with a bunch of voices go ahead and copy that go down to try it out paste it in here fix the slashes and execute easy here are some final examples of female and male voices using the UI that I have edited up in order to use this program if you want the UI program created using streamlit as well I will be hosting it on my patreon for my supporters now let me show you some examples so over here on the UI um on the left side we have English Chinese whatever I can't really read this this is Russian I'm pretty sure and then we can go ahead and click whatever speaker we have from the speaker folder that we set so there's EO there's these other two there's Tifa original there's Allen so Allen original sounds like a really deep and really deep voice man like a really cool n narrator we can click the convert button here and it'll take this input and turn it into a TTS I found out the food in your fridge are cooler decomposers so yeah there's alen and and here's alloy I'm pretty sure it's from a video game um yeah listen to this so it's going to go ahead and convert 2 seconds I found out the food in your fridge are cooler decomposes so that's actually really good I believe and honestly this for free and it doesn't even take up so many resources on your computer is really really cool so yeah thank you for watching and go ahead and leave a like subscribe for more future content I will be coming out with more stuff also bye-bye

Channel: Xavier

Views: 191

Keywords: coqui xtts, coqui, xtts, voice clone, text to speech, ai, tts, coqui tts, clone your voice, how to clone your voice ai, how to clone your voice free, rvc, retrieval based voice conversion, ai voice changer, ai voice changer app, ai voice changer free, ai voice changer mobile, ai voice generator, ai voice, artificial intelligence, ai voice cloning, voice cloning, voice cloning free, voice cloning ai, voice cloning tutorial

Id: kvxyymBMOY8

Length: 5min 32sec (332 seconds)

Published: Sat Feb 10 2024