RVC-Project / Retrieval-based-Voice-Conversion-WebUI

Easily train a good VC model with voice data <= 10 mins!
MIT License
22.73k stars 3.41k forks source link

PITY PITY PITY someone care about windows #2192

Open Soraiko opened 1 month ago

Soraiko commented 1 month ago

If someone successfully installs this on Windows PITY give instructions. I gave my time and mind already so maybe only money can solve this ? I followed instructions given in readme they're FUCKING NOT WORKING Python will you one day fuckign work on Windows FUCKING WORK ONE DAY

heartgold512 commented 1 month ago

i got it working but no gpu training as it isnt recognizing my gpu so im stuck in gpu training. so you need ffmpeg and ffprobe exe, virtualisation python environments .I assume you have a python version and its saved in environment variable. ie clone cd to clone py (version) -m venv (name) then name\Scripts\activate pip install torch torchvision torchaudio python -c "import torch; print(torch.version)" ie the c "something" is the output you need in my case it is cpu though some have c118 or c117 pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/(output ie this changes based on prev command)

pip install -r requirements.txt

python tools/download_models.py

you might need to launch with something.bat: @echo off call venv\Scripts\activate python app.py ? (i hope this helps) just make sure you have ffmpeg/ probe etc

heartgold512 commented 1 month ago

change go web.bat to CALL env\Scripts\activate python infer-web.py --port 7897 (or a port that works)