Learning to separate vocals from polyphonic mixtures via ensemble methods and structured output prediction.

Matt McVicar, Ral Santos-Rodriguez, Tijl De Bie

Browse the full ICASSP paper archive.