A Single Channel Speech Enhancement Using Time-frequency Sparsity

摘要：

For the problem Gaussian noise seriously interferes with speech signals, thts paper proposes a novel single channel speech enhancement algorithm using underdetermined blind signal separation (BSS) method. The algorithm builds a mathematical model which is similar to underdetermined BSS. It splits the time frequency domain into two areas: Joint areas (JA) and disjoints area (DA). The signals are thrown away in the JA. In the DA, it deletes the noise on the cross and applies the smooth filter to compensate the distortion. Finally, several speech signal experiments demonstrate that the algorithm performance and practical.

关键词： sparsity threshold speech enhancement underdetermined blind source separation disjointness

作者: Gongxian Sun Feng Gao Ming Xiao

作者单位: College of Electrics & Information Engineering,South China University of Technology, Guangzhou,China College of Electrics & Information Engineering,South China University of Technology, Guangzhou,China

会议类型: 国际会议

会议名称: 2011 International Conference on Communication and Electronics Information(ICCEI 2011)(2011年通信和电子信息国际会议)

会议地点: 海口

会议语种:英文

页码: 19-22

在线出版日期: 2011-02-22（万方平台首次上网日期，不代表论文的发表时间）