Intensity- and location-normalized training for HMM-based visual speech recognition.

Yoshihiko Nankaku, Keiichi Tokuda, Tadashi Kitamura

Browse the full Interspeech paper archive.