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.
I was looking in transformer.py and I don't see kv-cache implemented. It should speed up inference by quite a bit. I'm wondering if you guys are planning to do this.
Awesome repo, thanks for opensourcing it.
I was looking in transformer.py and I don't see kv-cache implemented. It should speed up inference by quite a bit. I'm wondering if you guys are planning to do this.
Thanks.