语音信号时间动态规正新方法

A Novel Dynamic Time Alignment Method for Speech Signals

利用函数逼近原理，提出了一种语音信号时间动态规正的新方法，并由此导出了可用于神经网络语音识别的新特征．重点讨论了时间动态规正的原理、新特征的提取方法和性质．实验及理论分析表明。

This paper presents a novel time alignment method based on the principle of function approximation. Using this method, we get a new feature which can be used in ANN based speech recognition. The fundamental principle of the new method, the extracting method and the characteristics of new feature are discussed. The experimental results and theoretical analysis show that the new feature has outstanding virtue in ANN based speech recognition and speech compression.