计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (29): 134-136.DOI: 10.3778/j.issn.1002-8331.2010.29.037

• 数据库、信号与信息处理 • 上一篇    下一篇

基于小波变换C0复杂度的语音端点检测方法

王纲金,赵 欢,胡 炼   

  1. 湖南大学 计算机与通信学院,长沙 410082
  • 收稿日期:2009-12-08 修回日期:2010-02-10 出版日期:2010-10-11 发布日期:2010-10-11
  • 通讯作者: 王纲金

Voice activity detection method based on wavelet transform C0 complexity

WANG Gang-jin,ZHAO Huan,HU Lian   

  1. School of Computer and Communication,Hunan University,Changsha 410082,China
  • Received:2009-12-08 Revised:2010-02-10 Online:2010-10-11 Published:2010-10-11
  • Contact: WANG Gang-jin

摘要: 对传统的C0复杂度语音端点检测方法改进,提出一种基于小波变换的C0复杂度(WC0)方法,其特征门限估计采用模糊C均值聚类算法和贝叶斯信息准则算法,并采用双门限法进行语音端点检测。在TIMIT连续语音库上的实验表明,在低信噪比环境下,WC0法的检测性能明显优于基于传统的C0复杂度法,特别是在车辆噪声和车内噪声环境下,WC0法表现出更好的检测性能。

Abstract: This paper proposes a Voice Activity Detection(VAD) method based on wavelet transform C0 complexity(WC0),which improves the traditional C0 complexity,using fuzzy C means clustering algorithm and Bayesian information criterion algorithm to estimate the thresholds of the WC0 characteristic,and using dual threshold method for VAD.Experiments on the TIMIT continuous speech database show that at low SNR environments,WC0 method is superior to C0 method.Especially in the vehicle noise and vehicle interior noise environments,WC0 method shows better detection performance.

中图分类号: