Speaker verification using Gaussian mixture model (GMM)

H. Hussain; S. H. Salleh; C. M. Ting; A. K. Ariff; I. Kamarulafizam; R. A. Suraya

doi:10.1007/978-3-642-21729-6_140

Speaker verification using Gaussian mixture model (GMM)

H. Hussain, S. H. Salleh, C. M. Ting, A. K. Ariff, I. Kamarulafizam, R. A. Suraya

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

4 Citations (Scopus)

Abstract

This paper applies GMM for SV on Malay speech. The speaker models were trained on maximum likelihood estimated. The system was evaluated with 23 client speakers with 56 imposters. Malay clean speech data was used. 20 training of 3.5s utterances are used. The best performance achieved using 256-Gaussian imposter model and 32-Gaussian client model gave 3.01% of EER.

Original language	English
Title of host publication	5th Kuala Lumpur International Conference on Biomedical Engineering 2011, BIOMED 2011
Pages	560-564
Number of pages	5
DOIs	https://doi.org/10.1007/978-3-642-21729-6_140
Publication status	Published - 2011
Externally published	Yes
Event	Kuala Lumpur International Conference on Biomedical Engineering (BIOMED) 2011 - Kuala Lumpur, Malaysia Duration: 20 Jun 2011 → 23 Jun 2011 Conference number: 5th https://link.springer.com/book/10.1007/978-3-642-21729-6 (Proceedings)

Publication series

Name	IFMBE Proceedings
Volume	35 IFMBE
ISSN (Print)	1680-0737

Conference

Conference	Kuala Lumpur International Conference on Biomedical Engineering (BIOMED) 2011
Abbreviated title	BIOMED 2011
Country/Territory	Malaysia
City	Kuala Lumpur
Period	20/06/11 → 23/06/11
Internet address	https://link.springer.com/book/10.1007/978-3-642-21729-6 (Proceedings)

Keywords

Equal Error Rate
Gaussian Mixture Model
Speaker Verification

Access to Document

10.1007/978-3-642-21729-6_140

Cite this

@inproceedings{c56eda5928e1463692559dd6fdfa21fe,

title = "Speaker verification using Gaussian mixture model (GMM)",

abstract = "This paper applies GMM for SV on Malay speech. The speaker models were trained on maximum likelihood estimated. The system was evaluated with 23 client speakers with 56 imposters. Malay clean speech data was used. 20 training of 3.5s utterances are used. The best performance achieved using 256-Gaussian imposter model and 32-Gaussian client model gave 3.01% of EER.",

keywords = "Equal Error Rate, Gaussian Mixture Model, Speaker Verification",

author = "H. Hussain and Salleh, {S. H.} and Ting, {C. M.} and Ariff, {A. K.} and I. Kamarulafizam and Suraya, {R. A.}",

note = "Copyright: Copyright 2011 Elsevier B.V., All rights reserved.; Kuala Lumpur International Conference on Biomedical Engineering (BIOMED) 2011, BIOMED 2011 ; Conference date: 20-06-2011 Through 23-06-2011",

year = "2011",

doi = "10.1007/978-3-642-21729-6_140",

language = "English",

isbn = "9783642217289",

series = "IFMBE Proceedings",

pages = "560--564",

booktitle = "5th Kuala Lumpur International Conference on Biomedical Engineering 2011, BIOMED 2011",

url = "https://link.springer.com/book/10.1007/978-3-642-21729-6",

}

Hussain, H, Salleh, SH, Ting, CM, Ariff, AK, Kamarulafizam, I & Suraya, RA 2011, Speaker verification using Gaussian mixture model (GMM). in 5th Kuala Lumpur International Conference on Biomedical Engineering 2011, BIOMED 2011. IFMBE Proceedings, vol. 35 IFMBE, pp. 560-564, Kuala Lumpur International Conference on Biomedical Engineering (BIOMED) 2011, Kuala Lumpur, Malaysia, 20/06/11. https://doi.org/10.1007/978-3-642-21729-6_140

TY - GEN

T1 - Speaker verification using Gaussian mixture model (GMM)

AU - Hussain, H.

AU - Salleh, S. H.

AU - Ting, C. M.

AU - Ariff, A. K.

AU - Kamarulafizam, I.

AU - Suraya, R. A.

N1 - Conference code: 5th

PY - 2011

Y1 - 2011

N2 - This paper applies GMM for SV on Malay speech. The speaker models were trained on maximum likelihood estimated. The system was evaluated with 23 client speakers with 56 imposters. Malay clean speech data was used. 20 training of 3.5s utterances are used. The best performance achieved using 256-Gaussian imposter model and 32-Gaussian client model gave 3.01% of EER.

AB - This paper applies GMM for SV on Malay speech. The speaker models were trained on maximum likelihood estimated. The system was evaluated with 23 client speakers with 56 imposters. Malay clean speech data was used. 20 training of 3.5s utterances are used. The best performance achieved using 256-Gaussian imposter model and 32-Gaussian client model gave 3.01% of EER.

KW - Equal Error Rate

KW - Gaussian Mixture Model

KW - Speaker Verification

UR - http://www.scopus.com/inward/record.url?scp=79960006697&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-21729-6_140

DO - 10.1007/978-3-642-21729-6_140

M3 - Conference Paper

AN - SCOPUS:79960006697

SN - 9783642217289

T3 - IFMBE Proceedings

SP - 560

EP - 564

BT - 5th Kuala Lumpur International Conference on Biomedical Engineering 2011, BIOMED 2011

T2 - Kuala Lumpur International Conference on Biomedical Engineering (BIOMED) 2011

Y2 - 20 June 2011 through 23 June 2011

ER -

Speaker verification using Gaussian mixture model (GMM)

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this