If you are having trouble installing so-vits. Click here! (Windows)

[dillfrescott]

I made a handy conda env file with all the requirements that I installed and got sovits working. Please use this env file with miniconda or conda to prepare a working environment!

I can be contacted at cross@dill.moe if you need any help!

https://tinyurl.com/2p8ayph4

Tested on main branch. As of: March 1st, 2023

Here is also a link to a compressed archive of the entire env straight from my machine. It (should) just work if you unarchive the folder and place it under miniconda3\envs\ then just run a simple: conda activate vits

https://icedrive.net/s/N7SvDAF8Yt9fgiV3X3Xh46Ww74AT

[yonhung]

I got this error:

Pip subprocess error:
ERROR: Ignored the following versions that require a different python version: 0.52.0 Requires-Python >=3.6,<3.9; 0.52.0rc3 Requires-Python >=3.6,<3.9
ERROR: Could not find a version that satisfies the requirement torch==1.13.1+cu116 (from versions: 1.7.1, 1.8.0, 1.8.1, 1.9.0, 1.9.1, 1.10.0, 1.10.1, 1.10.2, 1.11.0, 1.12.0, 1.12.1, 1.13.0, 1.13.1)
ERROR: No matching distribution found for torch==1.13.1+cu116

failed

CondaEnvException: Pip failed

How do I fix this?

[dillfrescott]

Hmmmm, I've gotten that error before as well. I think one of the best solutions is to remove the torch==1.13.1+cu116 dependency from the file, or whatever it takes to get it to not throw a critical error, and then install pytorch manually with the correct cuda version. Something like:

pip3 install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu116

I apologize, my file may be quite flawed as to the install process. But it should give you a good starting point.

[dillfrescott]

@yonhung If that still does not work, I am preparing you a 7zip file of the complete env from my machine. You should be able to unarchive it and put it under miniconda3\envs\ and just activate it.

It will take about an hour or so and it will be ready! I'm compressing it right now!

[dillfrescott]

@yonhung Here it is!

https://icedrive.net/s/N7SvDAF8Yt9fgiV3X3Xh46Ww74AT

Hopefully that works by just placing the unarchived folder in your envs file! Please let me know if it works!

[yonhung]

@dillfrescott After I activate the environment, can I enter the so-vits-svc directory and run the code? I try to run inference_main.py. But I got this error. Should I install this package?

 import librosa
ModuleNotFoundError: No module named 'librosa'

[yonhung]

@dillfrescott Thanks a lot! It worked now.😄 I just forgot to add python before inference_main.py.

[dillfrescott]

No problem! I'm glad it could help you out! Good luck with your voice conversions!