基于小波包的语音谱熵端点检测方法研究

端点检测是语音识别过程中的一个重要的环节,因此改善端点检测的效果一直是语音识别领域的一个重要课题。为了提高在背景噪声下语音信号端点检测的准确率,提出了一种基于小波包的谱熵端点检测方法。该方法对语音信号进行小波包变换,将每帧信号分解成多个子带,在此基础上计算每帧信号的子带能量,通过子带能量所占比例求出每帧信号的谱熵,最后确定新的门限值。仿真实验表明,该方法比传统方法更有效、更优越,能够比较准确地检测语音信号。

Method Research of Speech Spectral Entropy Endpoint Detection Based on Wavelet Packet

Endpoint detection is an important link in the speech recognition process,therefore,improving the effectiveness of Endpoint detection is an important subject in speech recognition field.In order to increase the accuracy of speech endpoint detection in background noise,a method research of spectral entropy endpoint detection based on wavelet packet is proposed.This method transforms speech signal with wavelet packet,each frame signal is decomposed into multiple sub-bands,on this basis sub-band energy of each frame signal is calculated,spectral entropy of each frame signal is computed through sub-band energy proportion,finally new threshold is determined.Simulation experiment shows that this method is more effective and more superior than traditional methods,which could comparative accuratly detect speech signal.