Audiocraft is a library for audio processing and generation with deep learning. It features the state-of-the-art EnCodec audio compressor / tokenizer, along with MusicGen, a simple and controllable music generation LM with textual and melodic conditioning.
MIT License
21.06k
stars
2.17k
forks
source link
Does audiocraft support torch.compile on MusicGen / AudioGen? #176
zeke-john
commented
8 months ago
I tried the MusciGen model with torch.compile but the resulting audio is cacaphonous without any musical integrity.
It works fine when I only decorate the encodec model functions like this with torchinductor: https://github.com/sspintel/audiocraft/commit/1a8ed5621eda49766499629b3740f1b5aced9f2a but when the transformer model functions are decorated, the resulting audio output is messy. Are there plans to add torch.compile support?