冲击噪声环境下基于信号子空间的多通道语音增强算法  

Speech enhancement approach based on signal subspace with multi-input in impulsive noise

作　　者：赵彦平[1] 赵晓晖[1] 顾海军[1] 

机构地区：[1]吉林大学通信工程学院,长春130012

出　　处：《吉林大学学报（工学版）》2007年第2期453-457,共5页Journal of Jilin University:Engineering and Technology Edition

基　　金：吉林省科技发展计划项目(20050521)

摘　　要：针对冲击噪声环境下的语音增强问题提出了一种不依赖任何语音信号模型、适用于冲击噪声的基于信号子空间分解的多通道语音增强算法。该算法把阵列信号处理中的协变异应用到语音增强中,对带噪语音信号的协变异系数矩阵进行特征值分解得到纯净语音信号的子空间。仿真实验结果表明:该算法对于冲击噪声有抑制作用,对高斯白噪声、高斯有色噪声也有一定的抑制效果。In order to solve the problem of speech enhancement in impulsive noise, on the basis of signal subspace decomposition without speech model, a multi-input speech enhancement algorithm was presented. Using covariation concept in array signal processing, the proposed algorithm can obtain clean speech signal subspace by eigen-decomposition of the covariation coefficient matrix of speech signal corrupted by impulsive noise. Simulation results show that the algorithm is effective to reduce the impacts of impulse noise, white Gaussian noise and Gaussian colored noise.

关 键 词：信息处理技术 语音增强 冲击噪声 协变异 信号子空间 

分 类 号：TN911.72[电子电信—通信与信息系统]