A self-calibrating algorithm for speaker tracking based on audio-visual statistical models.

Matthew J. Beal, Nebojsa Jojic, Hagai Attias

Browse the full ICASSP paper archive.