Speech Enhancement Based on Noise Eigenspace Projection

Dongwen YING  Masashi UNOKI  Xugang LU  Jianwu DANG 

Publication
IEICE TRANSACTIONS on Information and Systems  Vol.E92-D  No.5  pp.1137-1145
Publication Date: 2009/05/01
Online ISSN: 1745-1361
Print ISSN: 0916-8532
Type of Manuscript: PAPER
Category: Speech and Hearing
Keyword: 
speech enhancement,  noise eigenspace,  dimension reduction (DR),  Karhunen-Loeve transform (KLT), 

Full Text: PDF(384.7KB)

Summary: 
How to reduce noise with less speech distortion is a challenging issue for speech enhancement. We propose a novel approach for reducing noise with the cost of less speech distortion. A noise signal can generally be considered to consist of two components, a "white-like" component with a uniform energy distribution and a "color" component with a concentrated energy distribution in some frequency bands. An approach based on noise eigenspace projections is proposed to pack the color component into a subspace, named "noise subspace". This subspace is then removed from the eigenspace to reduce the color component. For the white-like component, a conventional enhancement algorithm is adopted as a complementary processor. We tested our algorithm on a speech enhancement task using speech data from the Texas Instruments and Massachusetts Institute of Technology (TIMIT) dataset and noise data from NOISEX-92. The experimental results show that the proposed algorithm efficiently reduces noise with little speech distortion. Objective and subjective evaluations confirmed that the proposed algorithm outperformed conventional enhancement algorithms.