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。
This paper proposes a Voice Activity Detection（VAD） method based on wavelet transform Co 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 WCo 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.
出处
《计算机工程与应用》
CSCD
北大核心
2010年第29期134-136,195,共4页
Computer Engineering and Applications
基金
湖南省自然科学基金重点项目(No.10JJ2046)
湖南省科技计划项目(No.05FJ3046)~~
关键词
语音端点检测
C0复杂度
小波变换
模糊C均值聚类算法
贝叶斯信息准则算法
voice activity detection
C0 complexity
wavelet transform
fuzzy C means clustering algorithm
Bayesian information criterion algorithm
