A Speech Enhancement Algorithm Based on Blind Signal Cancelation in Diffuse Noise Environments

Jaesik HWANG  Jaepil SEO  Ji-Won CHO  Hyung-Min PARK  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E99-A   No.1   pp.407-411
Publication Date: 2016/01/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E99.A.407
Type of Manuscript: LETTER
Category: Speech and Hearing
speech enhancement,  Wiener filtering,  beamforming,  blind signal extraction,  

Full Text: PDF(245.5KB)>>
Buy this Article

This letter describes a speech enhancement algorithm for stereo signals corrupted by diffuse noise. It estimates the noise signal and also a beamformed target signal based on blind target signal cancelation derived from sparsity minimization. Enhanced target speech is obtained by Wiener filtering using both the signals. Experimental results demonstrate the effectiveness of the proposed method.