search

lucidrains / audiolm-pytorch

Implementation of AudioLM, a SOTA Language Modeling Approach to Audio Generation out of Google Research, in Pytorch
MIT License
2.42k stars 261 forks source link

Can you please implement 'SingSong' by Google? #87

Closed adamfils2 closed 1 year ago

adamfils2 commented 1 year ago

SingSong generates backing instruments from input vocals. https://arxiv.org/abs/2301.12662v1 https://paperswithcode.com/paper/singsong-generating-musical-accompaniments Demo: https://storage.googleapis.com/sing-song/index.html

In the paper they use audiolm.

Could you please provide code to train this? I would really appreciate it. Thanks

lucidrains commented 1 year ago

Yea sure, closing because this is same as https://github.com/lucidrains/audiolm-pytorch/issues/86

mishav78 commented 3 months ago

did someone code the singsong implementation?

lucidrains commented 3 months ago

no i never got around to this

mishav78 commented 3 months ago

I will email you