Speech recognition using a wavelet packet adaptive network based fuzzy inference system

摘要

In this paper, an expert speech recognition system is presented. This paper especially deals with the combination of feature extraction and classification for real speech signals. A Wavelet packet adaptive network based fuzzy inference system (WPANFIS) model is developed in this study. WPANFIS consists of two layers: wavelet packet and adaptive network based fuzzy inference system. The wavelet packet layer is used for adaptive feature extraction in the time–frequency domain and is composed of wavelet packet decomposition and wavelet packet entropy. The performance of the developed system is evaluated by using noisy speech signals. Test results showing the effectiveness of the proposed speech recognition system are presented in the paper. The rate of correct classification is about 92% for the sample speech signals.