摘要
为了提高语音端点检测的适应性和鲁棒性,提出一种基于小波分析和支持向量机的语音端点检测算法。首先利用小波变换提取语音信号的特征量,然后将这些特征量作为支持向量机的输入进行训练和建模,最后判断出该信号的类别。仿真实验表明,相对于传统的语音端点检测算法,小波分析和支持向量机的检测算法提高了语音端点检测的正确率,有效降低了虚检率和漏检率,具有更好的适应性和鲁棒性,对不同信噪比的信号都有较好的检测能力。
In order to improve the adaptability and robustness of speech endpoint detection,this paper presented a algorithm for speech endpoint detection based on wavelet analysis and support vector machine.Firstly,the characteristic quantities of speech signals are obtained by the wavelet transformation.Then the input to support vector machine can be computed based on these characteristic quantities.Finally the signal＇s type can be determined.The simulation experiments results show that the proposed algorithm improves the detection rate,has better adaptability and robustness,and can detect signals with different SNR,compared with the traditional detection algorithms.
出处
《计算机科学》
CSCD
北大核心
2012年第6期244-246,265,共4页
Computer Science
基金
齐齐哈尔市科技局科技攻关基金项目(GYGG-09007)资助
关键词
小波分析
支持向量机
语音端点
特征提取
Wavelet analysis
Support vector machine
Speech Endpoints
Feature extraction
作者简介
朱恒军（1969-），男，硕士，副教授，主要研究方向为信号采集与信息处理、通信与信息系统，E-mail：hengjun_zhu@163．com。
