Generative model-based speaker clustering via mixture of von Mises-Fisher distributions.

Hao Tang, Stephen M. Chu, Thomas S. Huang

Browse the full ICASSP paper archive.