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 am confusing about the text2music config. It seems like in config/model/lm/default.yaml the cross_attention=False and I didn't see any overwrite in musicgen_lm.yaml but in the output of the fuser does include the original input and cross_attention_output. I am wondering whether the flag is set wrong or if I am missing something here. Thank you.
Confusing about the same question. self.cross_attention should have not executed when cross_attention=False but I don't know why the code ran successfully. Did you figure it out?
I am confusing about the text2music config. It seems like in
config/model/lm/default.yaml
thecross_attention=False
and I didn't see any overwrite inmusicgen_lm.yaml
but in the output of the fuser does include the originalinput
andcross_attention_output
. I am wondering whether the flag is set wrong or if I am missing something here. Thank you.