| 一种基于奇异谱的语音激活检测方法 |
| |
| 引用本文： | 曹亮,张天骐,周 圣,胡 然.一种基于奇异谱的语音激活检测方法[J].应用声学,2013,32(2):137-143. |
| |
| 作者姓名： | 曹亮 张天骐 周 圣 胡 然 |
| |
| 作者单位： | 重庆邮电大学 信号与信息处理重庆市重点实验室 重庆 400065;重庆邮电大学 信号与信息处理重庆市重点实验室 重庆 400065;重庆邮电大学 信号与信息处理重庆市重点实验室 重庆 400065;重庆邮电大学 信号与信息处理重庆市重点实验室 重庆 400065 |
| |
| 基金项目： | 国家自然科学基金项目(61071196, 61102131), 教育部新世纪优秀人才支持计划项目(NCET-10-0927), 信号与信息处理重庆市市级重点实验室建设项目(CSTC2009CA2003), 重庆市杰出青年基金项目(CSTC2011jjjq40002)和重庆市自然科学基金项目(CSTC2009BB2287, CSTC2010BB2398, CSTC2010BB2409, CSTC2010BB2411)资助。 |
| |
| 摘 要： | 为了提高语音激活检测在低信噪比环境中的检测性能，提出了一种基于奇异谱的语音激活检测方法。首先用多窗口方法计算每一帧语音信号的相关矩阵；然后对相关矩阵进行奇异值分解；利用奇异值可以反映有用信号和噪声分布情况的特性，将每一帧语音信号经过加权处理后的最大奇异值与自适应阈值进行比较进行语音激活检测。该方法原理简单，易于硬件实现，通过实验仿真表明，在低信噪比环境下，和基于对数能量方法相比，本文方法也能够很好的区分语音段和非语音段，有良好的检测性能。
|
| 关 键 词： | 语音激活检测，Slepian数据窗，离散扁椭圆序列，相关矩阵，奇异值分解，自适应阈值 |
| 收稿时间： | 8/2/2012 12:00:00 AM |
A method of voice activity detection based on spectrum of singular value |
| |
| CAO Liang,ZHANG Tianqi,ZHOU Sheng and HU Ran.A method of voice activity detection based on spectrum of singular value[J].Applied Acoustics,2013,32(2):137-143. |
| |
| Authors: | CAO Liang ZHANG Tianqi ZHOU Sheng and HU Ran |
| |
| Institution: | Chongqing Key Laboratory of Signal and Information Processing Chongqing University of Posts and Telecommunications, Chongqing 400065, China;Chongqing Key Laboratory of Signal and Information Processing Chongqing University of Posts and Telecommunications, Chongqing 400065, China;Chongqing Key Laboratory of Signal and Information Processing Chongqing University of Posts and Telecommunications, Chongqing 400065, China;Chongqing Key Laboratory of Signal and Information Processing Chongqing University of Posts and Telecommunications, Chongqing 400065, China |
| |
| Abstract: | In order to improve the performance of voice activation detection at low SNR(Signal to Noise Ratio), we proposed a detection approach of voice activity based on singular spectrum. Firstly, we calculate the correlation matrix for each frame of speech signal with multi-window approach; then performed singular value decomposition to the correlation matrix; due to the singular value reflects the characteristics of the useful signal and noise distribution, we can perform activity detection through comparing the weighted maximum singular value of each frame of speech signal with the adaptive threshold value. This method is simple and can be easily implemented in hardware. The simulation indicates that compared with energy method based on logarithm, in low SNR environment, this approach can better distinguish speech segments with non-voice segment better. |
| |
| Keywords: | Voice activity detection Slepian data window Discrete prolate spheroidal sequences Correlation matrix Singular value decomposition Adaptive threshold |
|
| 点击此处可从《应用声学》浏览原始摘要信息 |
| 点击此处可从《应用声学》下载免费的PDF全文 |
|
