A two-stage singing voice separation algorithm using spectro-temporal modulation features.

Frederick Z. Yen, Mao-Chang Huang, Tai-Shih Chi

Browse the full Interspeech paper archive.