so-vits-fork Installation Tutorial for Windows 10

[@CarlTheNPC01]

Every time I open the config (json) file, it keeps crashing. How do I fix this? More Info: Every time it crashes it says "values=list(hp.__dict__["spk"].keys()), set_to_index=0 KeyError: 'spk'"

[@Th4t-Ai-Guy]

Same

[@AIUndertheHood]

I loved how you got straight to the point! No bs. KZbin needs way more of this!!!

[@aHtoxbl4]

This is the most simple and useful guide I found... Thanks!

[@Engiscoud]

First tutorial guy that is actually explaining it simple and funny

[@bencesilling5862]

Bro is on point. No intrudoction no side talk just the essence 😂 anyways thanks!

[@PalpatineKZ]

Buddy your tutorial is amazing! I've seen few and this is most relevant. Thank you!

[@Cooly-D]

thanks for tutorial! I hope i can still open the GUI tomorrow .

[@eren95666]

yes very good video and simple. best video i have watched in decades (im 19)

[@unitedshoes2128]

Thank you it's the lowest quality video I've made in decades (I'm 20)

[@BowGiulius]

i still not understand where i can find the config.json file

[@POULETKFC]

same, i don"t find configs in the models

[@josecassimirojunior2980]

this is the true simple and perfectly useful guide! Thaks a lot!

[@andivax]

thank you very much! 1) please make a step-by-step video guide on Training. 2) how to setup Presets (there are GPU 1060 there)? 3) how to increase Quality (both processing and output)? 4) do you have Patreon? I want to donate.

[@arrowsplit8659]

Oh are you on a gtx 1060 like I am Does it work ok

[@unitedshoes2128]

1. I don't plan on training 2. Presets are most useful for changing real-time voice conversion settings on the fly. You don't need to worry about GPU when it comes to presets, just make sure the "Use GPU" box is checked 3. You need clean vocals for it to work well, any fuzziness or distortion causes artifacts/voice cracks in the output. So-vits also struggles when it comes to layered voices. I found that MDX Karaoke in UVR is really good at isolating single voices from layered vocals. Kim-Vocal is really good at giving you super clean isolations (UVR in description) 4. No Patreon, but cashapp is $kyletrentpng. I don't expect anything from anyone, I'm just glad to be helping people install this

[@hieuha3718]

@@unitedshoes2128 Mine is always out of CUDA memory error on GTX 1650, pls help

[@fdgfdgdfgdfgfdgdf]

@@hieuha3718 mininum is 6gb vram bro

[@kazuud2666]

i have a question. when doing the conda activate so-vits-fork and then doing >svcg. mine said: 'svcg' is not recognized as an internal or external command, operable program or batch file. i also tried doing just conda activate so-vits and it said the same thing, so i'm a bit lost and don't know where to go from here

[@asdkljmd1622]

I think you just have to run the "pip install -U so-vits-svc-fork" command again, it worked for me

[@maestrojuniorsousa]

Hello, I would like to know if this program makes it possible to clone the voice of singers to produce covers of songs and what price do you charge me to install on my computer

[@unitedshoes2128]

50 million dollar for kanye

[@foxjesus4411]

God-Tier tutorial. Thank you

[@alexgoesbananas]

I am missing the infer button, how can I fix it?

[@9ALiTY]

every video on this is so confusing

[@Sirbibi1308]

This was so helpful! Thank you so much!

[@jacobwright5469]

When I type svcg, I get this ‘svcg’ is not recognized as an internal or external command, operable program or batch file what could I do to fix?

[@supersaiyen1126]

me too

[@kyletrent.mp4]

Make sure you have python=3.10 when you create your environment, and make sure you are running the install commands in your environment. It should say your environment name on the left when you're inside it, and "base" when you're not inside an environment

[@kyletrent.mp4]

"python3 svcg" might work if you install everything outside an environment

[@XO-ir4jl]

when i put svcg, showing up this error (( from antlr4.error.Errors import LexerNoViableAltException, RecognitionException ModuleNotFoundError: No module named 'antlr4' )), what happing there ??

[@himari2059]

how to open it again if we close? ,anyone please.

[@adamramses9722]

Can you please make a video about training your own model via google colab ?

[@JJ-vp3bd]

did you find an answer

[@ironox5885]

Great tutorial my dude 👍

[@molobeats2880]

can't find the appropriate config file at 4:25 . someone plz help

[@FYL-TJ4ANVC]

Models without config file are meant to work with RVC not so-vits-fork

[@Hayday21465]

Man I love this guy

[@lucieduriez284]

Why do I have the same so-vits-fork page but without the "infer" key ??

[@faitheeuu]

Can someone help me? I downloaded the software, but when I go and search for the prompt.. I don’t see it (Mac)

[@blakehicks7205]

use terminal on your mac.

[@everyneed3959]

Do you have a tutorial on how to use the live ai voice filter?

[@pris1174]

did you find out how to use it?

[@beardox]

Hi! THank you! Do you know if there is a way to convert model to G_model? I want to use a voice that I found but I just have the model.pth version without json file

[@VaibhavShewale]

damn recording early morning

[@kyletrent.mp4]

This guy is a god

[@gokulk8014]

Thanks mate...it finally worked..!

[@straykidsurr1793]

how do i create a model?

[@ZARATE_LX]

Thank you, I have fun!! xD

[@TheLibrarianUSA]

THANK YOU SO MUCH! This helped me finally install SVCG

[@Xortfo]

Where do I get the Json configs? All I have are the index files.

[@wextiy]

did you find a solution?

[@Xortfo]

@wextiy nope just used a different build from the ai hub discord

[@wextiy]

@@Xortfo thanks

[@williamsjessem]

>conda create -n so-vits-fork python=3.10 ---- to make it easy to copy paste that into your anaconda prompt

[@h2rv]

Is there possibly a work around for AMD GPU users on windows or is it a strict no go, must use linux? Thanks for the vid as well

[@loyalhack]

Did you find an answer anywhere?

[@h2rv]

@@loyalhack Unfortunately not, I just ended up using the AI hub integrated one

[@loyalhack]

@@h2rv tysm for the reply

[@akhileshkannan8421]

How do I uninstall it and then reinstall it? When I type "svcg" into anacond powershell prompt, it doesnt recognise the term

[@asdkljmd1622]

I had the same problem and all I had to do was run the "pip install -U so-vits-svc-fork" command again

[@Rosetti.]

I'm getting this error code: "PermissionError: [Errno 13] Permission denied: '.' " Any idea what that is and how to fix it?

[@LIVEFROMHYPERSPACE]

Super helpful. Thanks

[@FuryisBACKHD]

Why when I turn on the program and some video, there is white noise in my headphones? So when I get out of the program and keep watching it, it continues. To fix it, I have to pull out the headphone plug and put it back in to fix it. Has anyone had problems with this?

[@perfredrikpov]

(base) C:\WINDOWS\system32>conda activate so-vits EnvironmentNameNotFound: Could not find conda environment: so-vits You can list all discoverable environments with `conda info --envs`. (base) C:\WINDOWS\system32>

[@unitedshoes2128]

Did you create an environment with conda create -n so-vits python=3.10

[@perfredrikpov]

@@unitedshoes2128 It works when i do the command "conda activate so-vits-fork"

[@Shhbbitsokay]

its the opposite for me. cant find so-vits-fork but can find so-vits. i did create an environment. (base) C:\Users\Yassin>conda activate so-vits-fork EnvironmentNameNotFound: Could not find conda environment: so-vits-fork You can list all discoverable environments with `conda info --envs`. (base) C:\Users\Yassin>conda info --envs # conda environments: # base * C:\Users\Yassin\anaconda3 so-vits C:\Users\Yassin\anaconda3\envs\so-vits

[@leoxmoon07]

when I type "svcg" it says that it doesn't exist and couldn't be find, any help?

[@leoxmoon07]

@@hich2774 not yet :(

[@leoxmoon07]

@@hich2774 tysm!!

[@OwlFatherTarnished]

hey, try putting in " conda activate so-vits-fork" instead of " conda activate so-vits" after that then try again with putting svcg. It should work good luck

[@thundershyofficial-2769]

for some reason when i train sometimes it doesent utilize my gpu and utilizes the cpu do you know whats going on when it was okay before ?

[@adaptiveaussie2216]

The models I download from hugging face don't have the config files, I am confused.

[@khalilbenalgia4279]

very comlexe what if i have installed it with bat file the self automatic installation ?

[@xpandacorex266]

i always get the permissionerror error 13... tried many tutorials

[@Theartisthenrycollins]

having trouble finding models to test out .. where can I find some celebs?

[@MrZuma569]

But can the cloned voice also be used in real time? That is, I sing and at the moment in real time the artist's cloned voice comes out of the mic?

[@amirrezamahboob]

Hello, I installed this software, but it doesn't have an infer button! to get output

[@yKuroZetsu]

where can i find some models?

[@frattuncbas]

How can i continue my pre-trained model later? I want to train it 20000+ epochs for most realistic quality but i need to run day by day my same pre trained model..Is it possible on Easygui Colab? I have Pro colab

[@Yamoyashi]

im getting an ErrnO 13 permission denied, whats up with that?

[@rinkaemina]

is it already 4.0 or i need to update? if i need update, how?

[@unicorn4010]

Please update this package regularly to get the latest features and bug fixes. pip install -U so-vits-svc-fork

[@Nathanizer]

Great tutorial, thanks. But I only got one problem. (might be just the settings, I think so.) But when i open the So-vits, my sound is just noise. The render might be good, but just when its open, it is white noise. Any help?

[@Sicki6]

every time svcg was not found or misspelled....

[@Yoshifire8]

So lets say I want to use the real time voice filter in a discord call. How do I set my microphone to be the live inference output? Virtual audio cable?

[@unitedshoes2128]

The only time I got it to work was using virtual audio cables and voicemeeter. It's really confusing to set up, I only did it a few times. Good luck lol

[@darksydeflow]

How do I train a custom model?

[@kagesatori]

I made everything as you did, but commands don't work. And where did you get from the "so-vits" env?

[@joon4470]

how to train own model

[@Kozekiyako]

i rewatch this like a few time and didnt get what i did wrong to get multiple errors but now i realized its me not watching the video carefully 💀

[@Tyosetioko]

i can't find any config.json file and model, how to fix it ?

[@unitedshoes2128]

You can grab model .zips from the AI Hub discord in the video description

[@alden3694]

Where does the file bounce to? Everytime I infer my vocals I hear the audio and I cant locate the file.

[@twizzyGen]

Is there a way to improve the output file? I don't know how ppl are making their output so much cleaner since we're using the same models.

[@unitedshoes2128]

Cleaner vocal stem probably. Try recording the verse with your voice in a good microphone and see how that comes out. Doc in the description has great information regarding Ultimate Vocal Remover (UVR)

[@north7295]

you have to record the audio with your own voice and try to sing it identical to the artist

[@mixey01]

I usually use Izotope RX to eliminate all unwanted artifacts. The dialogue de-erb section in RX cleans my output file very nicely

[@DESIKID_YT]

I have AMD with with window 10 what should I do ?

[@eosonera]

Do you know if there is a way to do batch inference?

[@mrperecz]

Hi! How can i convert wav file to pth file? Sorry im noob :)

[@frustasistumbleguys4900]

so how to train model of my self?

[@CRISISYoutube]

your the goat 😊

[@demetriadevonne-qc9sq]

i can't create so vits fork can u help me?

[@dennisbailey5663]

I just hear something similar to backwards vocal

[@SomedayTomorrow369]

Will this work for windows 11 ??

[@ATAQUITO]

Yes

[@doodledangernoodle2517]

Is it possible to make it do acapellas well?

[@prod.faccio9391]

Would this work on mac or only windows?

[@unitedshoes2128]

Windows/Linux only for Nvidia GPUs, Linux for AMD Cards (pytorch limitation)

[@Cooly-D]

Good you asked, I was about to try this on my m1 Mac, so it a no go then :( bummer

[@kostas9849]

Is it able to clone my own voice with this software from audio file?

[@unitedshoes2128]

Yup, that's what it does

[@kostas9849]

@@unitedshoes2128 does it support foreign languages?i want to clone a voice in greek

[@blovTV]

What if I don’t install anaconda will it still work?

[@cowwzymoo]

will anaconda 2019.10 work, later versions install hangs on me

[@cowwzymoo]

update: it does

[@abhiramika6112]

Do you know how to make your own model?

[@cowwzymoo]

​@@abhiramika6112nope srey

[@LuhNeeks007]

can someone send me an uzi model

[@roseannepark2370]

can I do it using Intel GPU?

[@thundershyofficial-2769]

also why did u say i hate myself at the end lol

[@tavionslaps]

Real time video?

[@prodovr]

thanks so much for this bro plus do you think you could send me the xxxtentacion voice model you have. thanks

[@MonDiabolique]

lol, i know this is gonna sound weird but you and I have a very similar voice. thought i was making this vid for a sec. did you AI model me?? haha

[@unitedshoes2128]

If you ever make a model from your own voice lmk, I tried to train my voice but I only have 6GB VRAM

[@John-gx7xe]

"I hate myself"🤣😂 - thanks

[@kauskaurabatz9639]

hi,30 dollars for a private tutorial..i just want to make multiple models and decide when to stop the training of the one i have....

[@nawafalfaisl7231]

بغيت اسم التطبيق

[@Devotee_of_void]

Like for heading right to the topic

[@yugoarchive]

thanks g

[@isaac10231]

lmfao the end of the tutorial. great job tho

[@kuzn1]

lost my shit @ 7:11 😂😂😂

[@zipzopbeesopdro]

Dont get anaconda 3

[@litgeeksquadofficial]

Thank you

[@TonysMusic1974]

I consider myself pretty computer literate but I have no idea what this dude is talking about 🤷‍♂️

[@chopov11]

FOR THE FUNNY

[@TagetesAlkesta]

What even is this lmao

[@unitedshoes2128]

Converting vocals to AI celebrity voices. I am Kanye

[@TagetesAlkesta]

@@unitedshoes2128 Ah thank you Kyle very cool.

[@SINGAQ]

Dosn't work. "Symbol not found __svml_cosf8_ha"

[@Jackwald001]

This was so helpful!!! Thank you!

[@wextiy]

Where do I find the Config Json ?

[@rivermonstermovie]

when I load the config.json file it kills the svcg application

[@bendy3096]

why does it say that i need to pay for pysimplegui? how do i fix it?