Automatic visual speech segmentation

Speech recognition techniques which rely on audio features of speech degrade in performance in noisy environments. Visual Speech Recognition helps this by incorporating a visual signal into the recognition process. The performance of automatic speech recognition (ASR) system can be significantly enhanced with additional information from visual speech elements such as the movement of lips, tongue, and teeth. This paper introduces a combined method for lip region extraction and mouth area estimation, which is then used to develop technique for automatic visual speech segmentation. The accuracy of this method is verified by applying it for syllable boundary separation and the following vowel segmentation in multi syllable words and phrases.