Analysis of Changes in Lip Movement for Utterance Content Identification

Abstract

An automatic minute generating system based on speech recognition technology can be used to improve the efficiency of minute taking in meetings. When only one microphone is used for speech recognition, the number of devices used is reduced, and the preparation cost and convenience improve. However, the speech recognition accuracy decreases due to the influence of noise as disadvantage. On the other hand, lip movements have characteristics that are unique to the uttered contents and can be acquired even in noisy environments. The utterance contents estimated from the lip movement can correct of voice recognition content. In this investigation, a lip movement feature for estimating utterances was analyzed, with the aim of improving the speech recognition accuracy of an automatic minute creation system.