Underdetermined Separation of Speech Mixture Based on Sparse Bayesian Learning. (17th October 2016)

• Record Type: Journal Article
• Authors: Wang, Zhe; Wang, Luyun; Li, Xiumei; Zhao, Lifan; Bi, Guoan
• Other Names: Feulvarch Eric Academic Editor.
• Is Part Of: Mathematical problems in engineering. Volume 2016(2016)
• Publication Date: 2016-10-17
• Subjects: Engineering mathematics -- Periodicals; 510.2462
• DOI: 10.1155/2016/3982360
• Languages: English
• ISSNs: 1024-123X

Abstract

Abstract : This paper describes a novel algorithm for underdetermined speech separation problem based on compressed sensing which is an emerging technique for efficient data reconstruction. The proposed algorithm consists of two steps. The unknown mixing matrix is firstly estimated from the speech mixtures in the transform domain by usingK -means clustering algorithm. In the second step, the speech sources are recovered based on an autocalibration sparse Bayesian learning algorithm for speech signal. Numerical experiments including the comparison with other sparse representation approaches are provided to show the achieved performance improvement.