Automatic Speech Reading by Oral Motion Tracking for User Authentication System

Automatic speech recognition (ASR) systems are used in recognizing speech with high accuracy rates. Visual information is important for human machine interface. It not only increases the accuracy of an Automatic Speech Recognition (ASR) but also improves its robustness. This paper presents an overview of different approaches used for speech recognition and concentrates on visual only lip reading system. Lip reading can be utilized in many applications such as hearing impaired aid and for noisy environment where speech is highly unrecognizable and as password entry system. The visual feature extraction methods are pixel based such as discrete cosine transform (DCT), discrete wavelet transform (DWT)etc. Other feature extraction methods utilize motion analysis of image sequences representing lip movement. This paper is a survey paper explaining comparisons, pros and cons, analysis of various techniques and methods for speech recognition by lip motion tracking.