作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2007, Vol. 33 ›› Issue (24): 35-36. doi: 10.3969/j.issn.1000-3428.2007.24.012

• 博士论文 • 上一篇    下一篇

基于独立分量分析的单通道语音增强算法

李鸿燕,赵菊敏,王华奎,萧宝瑾   

  1. 太原理工大学信息工程学院,太原030024
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-12-20 发布日期:2007-12-20

Single Channel Speech Enhancement Algorithm Based on Independent Component Analysis

LI Hong-yan, ZHAO Ju-min, WANG Hua-kui, XIAO Bao-jin   

  1. Institute of Information Engineering, Taiyuan University of Technology, Taiyuan 030024
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-12-20 Published:2007-12-20

摘要: 传统的独立分量分析要求观测信号的个数不能小于源信号的个数,无法直接对单路信号进行独立分量分析。为了能够利用独立分量分析分离加性噪声,须构造一路观测信号。基于语音信号的短时平稳的特性,该文提出一种构造噪声信号的算法,实现了信号与噪声的分离。仿真结果表明,利用该算法可得到很好的消噪结果,提高信号的信噪比。

关键词: 独立分量分析, 盲源分离, 语音增强, 单通道

Abstract: The standard independent component analysis algorithm require that the number of sensors is more than or equal to that of sources, so it is impossible to apply independent component analysis to a single channel signal directly. This paper proposes an algorithm for constructing a noise signal for noise reduction based on ICA, thereby noise and signal can be separated through ICA. Simulation result shows that much better de-noise effect and signal-noise ratio can be obtained by using this algorithm.

Key words: Independent Component Analysis(ICA), blind sources separation, speech enhancement, single channel

中图分类号: