fix: unprompted generation for musicgen

facebookresearch / audiocraft

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

20.18k stars 2.01k forks source link

fix: unprompted generation for musicgen #330

Closed piwell closed 8 months ago

piwell commented 8 months ago

The unprompted examples created by generate_audio aren't unprompted. Since the call for prompted and unprompted is the same both will do audio continuation. By setting prompted_duration = None for the unprompted case we will not do audio continuations. I also added an assert on the output to make sure this would be caught in the future.

piwell commented 8 months ago

See that this is fixed in: https://github.com/facebookresearch/audiocraft/pull/335