Robust speaker recognition using SNR-aware subspace-based enhancement and probabilistic SVMs

Jia Ching Wang; Jhing Fa Wang; Wai He Kuok; Hsiao Ping Lee; Chung Hsien Yang

doi:10.1109/ICME.2006.262742

Robust speaker recognition using SNR-aware subspace-based enhancement and probabilistic SVMs

Jia Ching Wang, Jhing Fa Wang, Wai He Kuok, Hsiao Ping Lee, Chung Hsien Yang

Department of Computer Science and Information Engineering

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

1 Scopus citations

Abstract

In this paper, we present a robust text-independent speaker recognition system. The proposed system mainly includes an SNR-aware subspace-based enhancement technique and probabilistic support vector machines (SVMs). First, we construct a perceptual filterbank from psycho-acoustic model and incorporate it with the subspace-based enhancement approach. The prior SNR of each subband within the perceptual filterbank is taken to decide the estimator's gain to effectively suppress environmental background noises. Next, this study uses probabilistic SVMs to identify the speaker from the enhanced speech. The superiority of the proposed system has been demonstrated by twenty speaker recognition from AURORA-2 database with in-car noises.

Original language	English
Title of host publication	2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings
Pages	1161-1164
Number of pages	4
DOIs	https://doi.org/10.1109/ICME.2006.262742
State	Published - 2006
Event	2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Toronto, ON, Canada Duration: 9 Jul 2006 → 12 Jul 2006

Publication series

Name	2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings
Volume	2006

Conference

Conference	2006 IEEE International Conference on Multimedia and Expo, ICME 2006
Country/Territory	Canada
City	Toronto, ON
Period	9/07/06 → 12/07/06

Access to Document

10.1109/ICME.2006.262742

Cite this

Wang, J. C., Wang, J. F., Kuok, W. H., Lee, H. P., & Yang, C. H. (2006). Robust speaker recognition using SNR-aware subspace-based enhancement and probabilistic SVMs. In 2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings (pp. 1161-1164). Article 4036811 (2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings; Vol. 2006). https://doi.org/10.1109/ICME.2006.262742

@inproceedings{32ebe8ae8cc642588f0256509fd5bb71,

title = "Robust speaker recognition using SNR-aware subspace-based enhancement and probabilistic SVMs",

abstract = "In this paper, we present a robust text-independent speaker recognition system. The proposed system mainly includes an SNR-aware subspace-based enhancement technique and probabilistic support vector machines (SVMs). First, we construct a perceptual filterbank from psycho-acoustic model and incorporate it with the subspace-based enhancement approach. The prior SNR of each subband within the perceptual filterbank is taken to decide the estimator's gain to effectively suppress environmental background noises. Next, this study uses probabilistic SVMs to identify the speaker from the enhanced speech. The superiority of the proposed system has been demonstrated by twenty speaker recognition from AURORA-2 database with in-car noises.",

author = "Wang, {Jia Ching} and Wang, {Jhing Fa} and Kuok, {Wai He} and Lee, {Hsiao Ping} and Yang, {Chung Hsien}",

year = "2006",

doi = "10.1109/ICME.2006.262742",

language = "???core.languages.en_GB???",

isbn = "1424403677",

series = "2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings",

pages = "1161--1164",

booktitle = "2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings",

note = "2006 IEEE International Conference on Multimedia and Expo, ICME 2006 ; Conference date: 09-07-2006 Through 12-07-2006",

}

Wang, JC, Wang, JF, Kuok, WH, Lee, HP & Yang, CH 2006, Robust speaker recognition using SNR-aware subspace-based enhancement and probabilistic SVMs. in 2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings., 4036811, 2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings, vol. 2006, pp. 1161-1164, 2006 IEEE International Conference on Multimedia and Expo, ICME 2006, Toronto, ON, Canada, 9/07/06. https://doi.org/10.1109/ICME.2006.262742

Robust speaker recognition using SNR-aware subspace-based enhancement and probabilistic SVMs. / Wang, Jia Ching; Wang, Jhing Fa; Kuok, Wai He et al.
2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings. 2006. p. 1161-1164 4036811 (2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings; Vol. 2006).

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

TY - GEN

T1 - Robust speaker recognition using SNR-aware subspace-based enhancement and probabilistic SVMs

AU - Wang, Jia Ching

AU - Wang, Jhing Fa

AU - Kuok, Wai He

AU - Lee, Hsiao Ping

AU - Yang, Chung Hsien

PY - 2006

Y1 - 2006

N2 - In this paper, we present a robust text-independent speaker recognition system. The proposed system mainly includes an SNR-aware subspace-based enhancement technique and probabilistic support vector machines (SVMs). First, we construct a perceptual filterbank from psycho-acoustic model and incorporate it with the subspace-based enhancement approach. The prior SNR of each subband within the perceptual filterbank is taken to decide the estimator's gain to effectively suppress environmental background noises. Next, this study uses probabilistic SVMs to identify the speaker from the enhanced speech. The superiority of the proposed system has been demonstrated by twenty speaker recognition from AURORA-2 database with in-car noises.

AB - In this paper, we present a robust text-independent speaker recognition system. The proposed system mainly includes an SNR-aware subspace-based enhancement technique and probabilistic support vector machines (SVMs). First, we construct a perceptual filterbank from psycho-acoustic model and incorporate it with the subspace-based enhancement approach. The prior SNR of each subband within the perceptual filterbank is taken to decide the estimator's gain to effectively suppress environmental background noises. Next, this study uses probabilistic SVMs to identify the speaker from the enhanced speech. The superiority of the proposed system has been demonstrated by twenty speaker recognition from AURORA-2 database with in-car noises.

UR - http://www.scopus.com/inward/record.url?scp=34247553323&partnerID=8YFLogxK

U2 - 10.1109/ICME.2006.262742

DO - 10.1109/ICME.2006.262742

M3 - 會議論文篇章

AN - SCOPUS:34247553323

SN - 1424403677

SN - 9781424403677

T3 - 2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings

SP - 1161

EP - 1164

BT - 2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings

T2 - 2006 IEEE International Conference on Multimedia and Expo, ICME 2006

Y2 - 9 July 2006 through 12 July 2006

ER -

Robust speaker recognition using SNR-aware subspace-based enhancement and probabilistic SVMs

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this