Corruptinator

Quote

assuming you have a dataset of isolated sources

Oh, I didn't realize that part. Sorry. Seems that Spleeter requires more to make itself work. Thanks for clarifying Jim. While Spleeter may not seem useful it just got me thinking about how A.I. is or can be implemented in audio editing. Numerous possibilities.

Just not too long ago Deezer open sourced this interesting A.I. software called Spleeter. Its a A.I. tool trained and designed to separate audio based on vocals or instruments. Its not by any means perfect but a facinating tool that could be potentially implemented into cakewalk:

https://github.com/deezer/spleeter

Spleeter, a Deezer source seperation library is written in python and uses Tensorflow to separate (According to their Github Readme):

Vocals (singing voice) / accompaniment separation (2 stems)

Vocals / drums / bass / other separation (4 stems)

Vocals / drums / bass / piano / other separation (5 stems)

Pretty cool A.I. software to be honest. Would anybody here agree that this should be implemented into Cakewalk?