Searching-and-averaging method of underdetermined blind speech signal separation in time domain  

作　　者：XIAO Ming XIE ShengLi FU YuLi 

机构地区：[1]School of Electrics & Information Engineering, South China University of Technology, Guangzhou 510641, China [2]Department of Electrics & Information Engineering, Maoming College, Maoming 525000, China

出　　处：《Science in China(Series F)》2007年第5期771-782,共12页中国科学（F辑英文版）

基　　金：Supported by the National Natural Science Foundation of China (Grant Nos. U0635001, 60505005 and 60674033);the Natural Science Fund of Guangdong Province (Grant Nos. 04205783 and 05006508);the Specialized Prophasic Basic Research Projects of the Ministry of Science and Technology of China (Grant No. 2005CCA04100)

摘　　要：Underdetermined blind signal separation （BSS） （with fewer observed mixtures than sources） is discussed. A novel searching-and-averaging method in time domain （SAMTD） is proposed. It can solve a kind of problems that are very hard to solve by using sparse representation in frequency domain. Bypassing the disadvantages of traditional clustering （e.g., K-means or potential-function clustering）, the durative- sparsity of a speech signal in time domain is used. To recover the mixing matrix, our method deletes those samples, which are not in the same or inverse direction of the basis vectors. To recover the sources, an improved geometric approach to overcomplete ICA （Independent Component Analysis） is presented. Several speech signal experiments demonstrate the good performance of the proposed method.Underdetermined blind signal separation （BSS） （with fewer observed mixtures than sources） is discussed. A novel searching-and-averaging method in time domain （SAMTD） is proposed. It can solve a kind of problems that are very hard to solve by using sparse representation in frequency domain. Bypassing the disadvantages of traditional clustering （e.g., K-means or potential-function clustering）, the durative- sparsity of a speech signal in time domain is used. To recover the mixing matrix, our method deletes those samples, which are not in the same or inverse direction of the basis vectors. To recover the sources, an improved geometric approach to overcomplete ICA （Independent Component Analysis） is presented. Several speech signal experiments demonstrate the good performance of the proposed method.

关 键 词：underdetermined blind signal separation sparse representation searching-and-averaging method overcomplete independent component analysis 

分 类 号：TN911.2[电子电信—通信与信息系统]