Update requirements.txt

[oxysoft]

Support latest version of torchaudio (already functional with no changes)

[oxysoft]

I had some problems using demucs in my project / cloud compute which uses torch 2.3.1, however I have updated all the libraries 2.3.1 and kept demucs and it's working just fine. Not all features have been tested, so keep that in mind

[CarlGao4]

Actually, the global backend selector has been completely removed since torchaudio 2.2. Users can't switch audio backend when separating, so the separation won't be affected using torchaudio 2.2+. However, training requires this feature or passing the desire audio backend would make the code mess. Besides, deciding how to specify the backend according to torchaudio version will make the logic even complex, so I think limiting torchaudio versions to below 2.2 is a good choice.

[str-it]

thx. could not install without that fix

[samuelbradshaw]

If I'm understanding the README correctly, requirements.txt is the requirements file for training, and requirements_minimal.txt is the requirements file for simple stem separation. If training requires torchaudio<2.2, then requirements.txt can't be safely updated. But it should be safe to update requirements_minimal.txt. I opened a new pull request for this: #9

[ROBERT-MCDOWELL]

I'm using demucs with torch 2.7.1 and seems to work.