Robust HMM-based speech/music segmentation.

Jitendra Ajmera, Iain McCowan, Herv Bourlard

Browse the full ICASSP paper archive.