Signal subspace approach for speech enhancement in nonstationary noises

Chiung Wen Li; Sheau-Fang Lei

doi:10.1109/ISCIT.2007.4392269

Signal subspace approach for speech enhancement in nonstationary noises

Chiung Wen Li, Sheau-Fang Lei

Department of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

3 Citations (Scopus)

Abstract

In this paper, we propose a speech enhancement algorithm that is based on signal subspace approach (SSA) combined with RL noise estimation for nonstationary noises. A signal/noise Karhunen-Loeve transform (SNK) is used to implement signal subspace decomposition for noisy speech signal. RL noise estimation is used for adaptive estimates over time for nonstationary noises. Since the noise covariance matrix for SSA can be recursively updated by RL noise estimation, it results that the proposed algorithm can efficiently suppress the noise while reducing speech distortion. Objective experimental measurements indicated that the proposed algorithm was preferred over other SSA methods when tested with sentences corrupted by nonstationary noises.

Original language	English
Title of host publication	ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings
Pages	1580-1585
Number of pages	6
DOIs	https://doi.org/10.1109/ISCIT.2007.4392269
Publication status	Published - 2007 Dec 1
Event	ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies - Sydney, Australia Duration: 2007 Oct 16 → 2007 Oct 19

Publication series

Name	ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings

Other

Other	ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies
Country	Australia
City	Sydney
Period	07-10-16 → 07-10-19

All Science Journal Classification (ASJC) codes

Computer Networks and Communications
Communication

Access to Document

10.1109/ISCIT.2007.4392269

Fingerprint Dive into the research topics of 'Signal subspace approach for speech enhancement in nonstationary noises'. Together they form a unique fingerprint.

Cite this

Li, C. W., & Lei, S-F. (2007). Signal subspace approach for speech enhancement in nonstationary noises. In ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings (pp. 1580-1585). [4392269] (ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings). https://doi.org/10.1109/ISCIT.2007.4392269

@inproceedings{f34c0c7f965e43b8bdb50c0239ab7ac5,

title = "Signal subspace approach for speech enhancement in nonstationary noises",

abstract = "In this paper, we propose a speech enhancement algorithm that is based on signal subspace approach (SSA) combined with RL noise estimation for nonstationary noises. A signal/noise Karhunen-Loeve transform (SNK) is used to implement signal subspace decomposition for noisy speech signal. RL noise estimation is used for adaptive estimates over time for nonstationary noises. Since the noise covariance matrix for SSA can be recursively updated by RL noise estimation, it results that the proposed algorithm can efficiently suppress the noise while reducing speech distortion. Objective experimental measurements indicated that the proposed algorithm was preferred over other SSA methods when tested with sentences corrupted by nonstationary noises.",

author = "Li, {Chiung Wen} and Sheau-Fang Lei",

year = "2007",

month = dec,

day = "1",

doi = "10.1109/ISCIT.2007.4392269",

language = "English",

isbn = "1424409772",

series = "ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings",

pages = "1580--1585",

booktitle = "ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings",

note = "ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies ; Conference date: 16-10-2007 Through 19-10-2007",

}

Li, CW & Lei, S-F 2007, Signal subspace approach for speech enhancement in nonstationary noises. in ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings., 4392269, ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings, pp. 1580-1585, ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies, Sydney, Australia, 07-10-16. https://doi.org/10.1109/ISCIT.2007.4392269

Signal subspace approach for speech enhancement in nonstationary noises. / Li, Chiung Wen; Lei, Sheau-Fang.

ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings. 2007. p. 1580-1585 4392269 (ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Signal subspace approach for speech enhancement in nonstationary noises

AU - Li, Chiung Wen

AU - Lei, Sheau-Fang

PY - 2007/12/1

Y1 - 2007/12/1

N2 - In this paper, we propose a speech enhancement algorithm that is based on signal subspace approach (SSA) combined with RL noise estimation for nonstationary noises. A signal/noise Karhunen-Loeve transform (SNK) is used to implement signal subspace decomposition for noisy speech signal. RL noise estimation is used for adaptive estimates over time for nonstationary noises. Since the noise covariance matrix for SSA can be recursively updated by RL noise estimation, it results that the proposed algorithm can efficiently suppress the noise while reducing speech distortion. Objective experimental measurements indicated that the proposed algorithm was preferred over other SSA methods when tested with sentences corrupted by nonstationary noises.

AB - In this paper, we propose a speech enhancement algorithm that is based on signal subspace approach (SSA) combined with RL noise estimation for nonstationary noises. A signal/noise Karhunen-Loeve transform (SNK) is used to implement signal subspace decomposition for noisy speech signal. RL noise estimation is used for adaptive estimates over time for nonstationary noises. Since the noise covariance matrix for SSA can be recursively updated by RL noise estimation, it results that the proposed algorithm can efficiently suppress the noise while reducing speech distortion. Objective experimental measurements indicated that the proposed algorithm was preferred over other SSA methods when tested with sentences corrupted by nonstationary noises.

UR - http://www.scopus.com/inward/record.url?scp=46749116867&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=46749116867&partnerID=8YFLogxK

U2 - 10.1109/ISCIT.2007.4392269

DO - 10.1109/ISCIT.2007.4392269

M3 - Conference contribution

AN - SCOPUS:46749116867

SN - 1424409772

SN - 9781424409778

T3 - ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings

SP - 1580

EP - 1585

BT - ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies Proceedings

T2 - ISCIT 2007 - 2007 International Symposium on Communications and Information Technologies

Y2 - 16 October 2007 through 19 October 2007

ER -