English

应用科学学报

• 论文 • 上一篇    下一篇

窄带噪声下的子空间语音增强算法

陈国明 赵力 邹采荣   

  1. 东南大学 无线电工程系 江苏 南京 210096
  • 收稿日期:2006-06-06 修回日期:2006-12-01 出版日期:2007-05-31 发布日期:2007-05-31

Speech Enhancement Based on Subspace Method for Narrowband Noise

CHEN Guo－ming, ZHAO Li, ZOU Cai－rong   

  1. Department of Radio Engineering, Southeast University, Nanjing 210096，China
  • Received:2006-06-06 Revised:2006-12-01 Online:2007-05-31 Published:2007-05-31

被引次数

9

摘要： 传统的子空间语音增强方法在推导过程中，一般假定噪声的方差矩阵是满秩的.在窄带噪声或者音调噪声下这一条件就不能满足。本文采用直接对噪声信号进行子空间白化的方法，将传统的子空间语音增强方法推广到了窄带噪声的情况。仿真结果表明了该算法的有效性。

关键词: 语音增强, 子空间, 降秩

Abstract: A single channel speech enhancement method for noisy speech signals is proposed based on the subspace method. Conditional subspace based enhancement methods require that the noise covariance matrix is full-ranked. Unfortunately this is not satisfied in case of narrowband noise or tonal noise. The proposed method directly deals with rank-deficient noise using a subspace whitening method. Experimental results demonstrate the effectiveness of the proposed algorithms.

Key words: speech enhancement, subspace, rank-deficient