Visual speech recognition using artificial neural networking

Abstract

Automatic Speech Recognition plays an important role in human-computer interaction, which can be applied in various application like crime-fighting and helping the hearing-impaired consists of two domain – Audio Speech Recognition and Visual Speech Recognition. This thesis is based on Recognition of Speech in the visual domain only. This paper provides a new approach to lip reading Bengali words using a combination of the curvature of the inner and outer lips and Neural Networks. The method uses a more robust a faster algorithm to detect the lip contour than conventional methods used so far. Processing multiple frames and by collecting the contours, we can predict the Bengali words that are stored inside the database. Our thesis will mainly focus on detecting some specific Bengali words.