Speech Recognition Based on Dynamic and Static Feature Parameter Integration
基于动静态组合特征参数的语音识别

In this paper,we propose a new speech recognition method of dynamic and static feature integration.Wavelet packet transformation method is introduced to feature parameters in virtue of MFCC(Mel-Frequency Cepstrum Coefficient),then combined difference feature formed dynamic and static feature parameter integration.The simulation results indicate that the new feature parameter is better than MFCC,that is a very nice speech feature parameter.At the same time,chaotic dynamics is introduced to neurons,and a chao...