The MATLAB Simulation for More Generic Speech Recognizing Chinese Isolated Word System
Guohui Wang, Yong Guan, Yan Zhang · 2009
This paper introduces the basic theory of speech recognition, including speech signal pre-emphasis, endpoint detection and feature extraction, pattern recognition. On this basis, we introduce the MFCC (Mel-frequency cepstral coefficients) specific extraction method. Secondly, from the recognition rate and improve the speed of approach to identify proposed DTW (dynamic time warping) algorithm to improve the research, in order to achieve a more satisfactory effect of speech recognition. Experimental results show that the sets of speech recognition system to identify the model to meet the general conditions of the application.