Discriminative training of HMM stream exponents for audio-visual speech recognition.

Gerasimos Potamianos, Hans Peter Graf

Browse the full ICASSP paper archive.