Possible to train AI on instrumental of a track to isolate vocals? #1580
henrikwils
started this conversation in
General
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hey all
UVR is pretty damn good, but not perfect.
A lot of tracks are released with an instrumental, and even though the general consensus is that you can "just" invert those, it rarely works (and the lack of good acapellas show that as well).
I was thinking, if perhaps it would be possible to train an AI model on just the instrumental track that you want to extract the vocals from and then use that model to isolate the vocals?
Beta Was this translation helpful? Give feedback.
All reactions