基于小波神经网络的语音端点检测算法  

Speech endpoints detection method based on wavelet neural network

作　　者：胡伟[1] 郑明才[1] 

机构地区：[1]湖南第一师范学院科研处,长沙410002

出　　处：《计算机工程与应用》2013年第12期191-194,共4页Computer Engineering and Applications

基　　金：湖南省高校科技创新团队支持计划资助(湘教通[2010]212号);湖南省教育厅科学研究项目(No.11C0280)

摘　　要：为了提高语音端点检测效果,将小波分析和神经网络相融合,提出一种基于小波神经网络的语音端点检测算法(WA-PCA-RBF)。利用小波分析提取语音信号的特征向量,采用主成分分析法选择语音信号特征,消除冗余特征,将选择特征向量作为RBF神经网络输入,通过遗传算法优化RBF神经网络参数建立语音端检测模型。结果表明,相对于传统语音端点检测算法,WA-PCA-RBF提高了语音端点检测正确率,具有更好的适应性和鲁棒性,可满足实际系统需求。In order to improve the adaptability and robustness of speech endpoint detection, this paper proposes a speech end- point detection method based on wavelet analysis and neural network （WA-PCA-RBF）. The features of speech signals are ex- tracted by wavelet analysis; the features are selected by Principal Component Analysis to remove redundant features; the select- ed features are input to RBF neural network to build the speech endpoints detection model in which the RBF neural network＇ s parameters are optimized by genetic algorithm. The results show that the proposed method has improved the detection rate, and it has better adaptability and robustness in complicated noise environment compared with the traditional detection methods.

关 键 词：小波分析 神经网络 语音端点 特征提取 特征选择 

分 类 号：TN91[电子电信—通信与信息系统]