Separator Model
This model separates multiple voices within a recording. It's useful for cases like extracting the main vocals from a track with layered vocals / harmonies
