Word Recognition Using Adaptive Mel-Cepstral Analysis

Tetsuya Hoya, Hitoshi Ihara, Takamitsu Asano, Yoshihisa Ishida

Abstract: Neural networks have played an important role in many fields of signal processing. One of the fruitful application using neural networks is the development of speech recognition system. In this paper, a practical word recognition system using the technique of adaptive mel-cepstral analysis is presented. The experimental results show the proposed method is suitable for the practical word recognition system.