HMM-based audio-visual speech recognition integrating geometric and appearance-based visual features.

Michael T. Chan

Browse the full MMSP paper archive.