We present two models (issued from psychology) of audiovisual integration for speech perception. We propose an implementation of these models for audiovisual vowel recognition and we compare the results of these models in relation to their efficiency for speech recognition in noise.
Keywords: audiovisual speech recognition and perception.