Blind Source Separation Based on a Beamformer Array and Time Frequency Binary Masking

Author

Cermak, Jan ; Araki, Shoko ; Sawada, Hiroshi ; Makino, Shoji

Author_Institution

NTT Commun. Sci. Lab., NTT Corp., Kyoto

Volume

1

fYear

2007

fDate

15-20 April 2007

Abstract

This paper deals with a new technique for blind source separation (BSS) from convolutive mixtures. We present a three-stage separation system employing time-frequency binary masking, beamforming and a non-linear post processing technique. The experiments show that this system outperforms conventional time-frequency binary masking (TFBM) in both (over-)determined and underdetermined cases. Moreover it removes the musical noise and reduces interference in time-frequency slots extracted by TFBM.

Keywords

array signal processing; blind source separation; speech intelligibility; speech processing; time-frequency analysis; beamformer array; blind source separation; convolutive mixtures; interference reduction; musical noise; nonlinear post processing technique; three-stage separation system; time frequency binary masking; time-frequency slot extraction; Array signal processing; Blind source separation; Interference; Laboratories; Noise reduction; Signal processing; Source separation; Speech enhancement; Speech processing; Time frequency analysis; Array signal processing; speech enhancement; speech processing;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 2007. ICASSP 2007. IEEE International Conference on

Conference_Location

Honolulu, HI

ISSN

1520-6149

Print_ISBN

1-4244-0727-3

Electronic_ISBN

1520-6149

Type

conf

DOI

10.1109/ICASSP.2007.366637

Filename

4217037