Removal of residual cross-talk components in blind source separation using time-delayed spectral substration

Ryo Mukai; Shoko Araki; Hiroshi Sawada; Shoji Makino

Removal of residual cross-talk components in blind source separation using time-delayed spectral substration

Ryo Mukai^*, Shoko Araki, Hiroshi Sawada, Shoji Makino

^*Corresponding author for this work

Research output: Contribution to journal › Conference article › peer-review

19 Citations (Scopus)

Abstract

This paper describes a post processing method to refine output signals obtained by Blind Source Separation (BSS). The performance of BSS using Independent Component Analysis (ICA) declines significantly in a reverberant environment. The degradation is mainly caused by the cross-talk components derived from the reverberation of the jammer signal. Utilizing this knowledge, we propose a new method, time-delayed non-stationary spectral substraction, which removes the residual components from the separated signals precisely. The proposed method compensates for the weakness of BSS in a reverberant environment. Experimental results using speech signals show that the proposed method improves the signal-to-noise ratio by 3 to 5 dB.

Original language	English
Pages (from-to)	II/1789-II/1792
Journal	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2
Publication status	Published - 2002
Externally published	Yes
Event	2002 IEEE International Conference on Acoustic, Speech and Signal Processing - Orlando, FL, United States Duration: 2002 May 13 → 2002 May 17

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Cite this

Removal of residual cross-talk components in blind source separation using time-delayed spectral substration. / Mukai, Ryo; Araki, Shoko; Sawada, Hiroshi et al.
In: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Vol. 2, 2002, p. II/1789-II/1792.

Research output: Contribution to journal › Conference article › peer-review

@article{34f67f26058e45fda7c42723d54a8918,

title = "Removal of residual cross-talk components in blind source separation using time-delayed spectral substration",

abstract = "This paper describes a post processing method to refine output signals obtained by Blind Source Separation (BSS). The performance of BSS using Independent Component Analysis (ICA) declines significantly in a reverberant environment. The degradation is mainly caused by the cross-talk components derived from the reverberation of the jammer signal. Utilizing this knowledge, we propose a new method, time-delayed non-stationary spectral substraction, which removes the residual components from the separated signals precisely. The proposed method compensates for the weakness of BSS in a reverberant environment. Experimental results using speech signals show that the proposed method improves the signal-to-noise ratio by 3 to 5 dB.",

author = "Ryo Mukai and Shoko Araki and Hiroshi Sawada and Shoji Makino",

year = "2002",

language = "English",

volume = "2",

pages = "II/1789--II/1792",

journal = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

issn = "1520-6149",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "2002 IEEE International Conference on Acoustic, Speech and Signal Processing ; Conference date: 13-05-2002 Through 17-05-2002",

}

TY - JOUR

T1 - Removal of residual cross-talk components in blind source separation using time-delayed spectral substration

AU - Mukai, Ryo

AU - Araki, Shoko

AU - Sawada, Hiroshi

AU - Makino, Shoji

PY - 2002

Y1 - 2002

N2 - This paper describes a post processing method to refine output signals obtained by Blind Source Separation (BSS). The performance of BSS using Independent Component Analysis (ICA) declines significantly in a reverberant environment. The degradation is mainly caused by the cross-talk components derived from the reverberation of the jammer signal. Utilizing this knowledge, we propose a new method, time-delayed non-stationary spectral substraction, which removes the residual components from the separated signals precisely. The proposed method compensates for the weakness of BSS in a reverberant environment. Experimental results using speech signals show that the proposed method improves the signal-to-noise ratio by 3 to 5 dB.

AB - This paper describes a post processing method to refine output signals obtained by Blind Source Separation (BSS). The performance of BSS using Independent Component Analysis (ICA) declines significantly in a reverberant environment. The degradation is mainly caused by the cross-talk components derived from the reverberation of the jammer signal. Utilizing this knowledge, we propose a new method, time-delayed non-stationary spectral substraction, which removes the residual components from the separated signals precisely. The proposed method compensates for the weakness of BSS in a reverberant environment. Experimental results using speech signals show that the proposed method improves the signal-to-noise ratio by 3 to 5 dB.

UR - http://www.scopus.com/inward/record.url?scp=17444451815&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=17444451815&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:17444451815

SN - 1520-6149

VL - 2

SP - II/1789-II/1792

JO - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

JF - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

T2 - 2002 IEEE International Conference on Acoustic, Speech and Signal Processing

Y2 - 13 May 2002 through 17 May 2002

ER -

Removal of residual cross-talk components in blind source separation using time-delayed spectral substration

Abstract

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this