The effect of GMM order and CMS on speaker recognition with reverberant speech

Noam R. Shabtai; Yaniv Zigel; Boaz Rafaely

doi:10.1109/HSCMA.2008.4538707

The effect of GMM order and CMS on speaker recognition with reverberant speech

Noam R. Shabtai, Yaniv Zigel, Boaz Rafaely

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

7 Scopus citations

Abstract

Speaker recognition is used today in a wide range of applications. The presence of reverberation, in hands-free systems for example, results in performance degradation. The effect of reverberation on the feature vectors and its relation to optimal GMM order are investigated. Optimal model order is calculated in terms of minimum BIC and KIC, and tested for EER of a GMM-based speaker recognition system. Experimental results show that for high reverberation time, reducing model order reduces EER values of speaker recognition. The effect of CMS on state of the art GMM and AGMM-based speaker recognition systems is investigated for reverberant speech. Results show that high reverberation time reduces the effectiveness of CMS.

Original language	English
Title of host publication	2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008
Pages	144-147
Number of pages	4
DOIs	https://doi.org/10.1109/HSCMA.2008.4538707
State	Published - 3 Sep 2008
Event	2008 Hands-free Speech Communication and Microphone Arrays, HSCMA 2008 - Trento, Italy Duration: 6 May 2008 → 8 May 2008

Publication series

Name	2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008

Conference

Conference	2008 Hands-free Speech Communication and Microphone Arrays, HSCMA 2008
Country/Territory	Italy
City	Trento
Period	6/05/08 → 8/05/08

Keywords

Cepstral mean subtraction
Gaussian mixture model
Model order
Reverberation
Speaker recognition

ASJC Scopus subject areas

Hardware and Architecture
Electrical and Electronic Engineering
Communication

Access to Document

10.1109/HSCMA.2008.4538707

Cite this

Shabtai, N. R., Zigel, Y., & Rafaely, B. (2008). The effect of GMM order and CMS on speaker recognition with reverberant speech. In 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008 (pp. 144-147). Article 4538707 (2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008). https://doi.org/10.1109/HSCMA.2008.4538707

@inproceedings{03ea532ece0d42f4b034c3f32e6d7311,

title = "The effect of GMM order and CMS on speaker recognition with reverberant speech",

abstract = "Speaker recognition is used today in a wide range of applications. The presence of reverberation, in hands-free systems for example, results in performance degradation. The effect of reverberation on the feature vectors and its relation to optimal GMM order are investigated. Optimal model order is calculated in terms of minimum BIC and KIC, and tested for EER of a GMM-based speaker recognition system. Experimental results show that for high reverberation time, reducing model order reduces EER values of speaker recognition. The effect of CMS on state of the art GMM and AGMM-based speaker recognition systems is investigated for reverberant speech. Results show that high reverberation time reduces the effectiveness of CMS.",

keywords = "Cepstral mean subtraction, Gaussian mixture model, Model order, Reverberation, Speaker recognition",

author = "Shabtai, {Noam R.} and Yaniv Zigel and Boaz Rafaely",

year = "2008",

month = sep,

day = "3",

doi = "10.1109/HSCMA.2008.4538707",

language = "English",

isbn = "9781424423385",

series = "2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008",

pages = "144--147",

booktitle = "2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008",

note = "2008 Hands-free Speech Communication and Microphone Arrays, HSCMA 2008 ; Conference date: 06-05-2008 Through 08-05-2008",

}

Shabtai, NR, Zigel, Y & Rafaely, B 2008, The effect of GMM order and CMS on speaker recognition with reverberant speech. in 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008., 4538707, 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008, pp. 144-147, 2008 Hands-free Speech Communication and Microphone Arrays, HSCMA 2008, Trento, Italy, 6/05/08. https://doi.org/10.1109/HSCMA.2008.4538707

The effect of GMM order and CMS on speaker recognition with reverberant speech. / Shabtai, Noam R.; Zigel, Yaniv ; Rafaely, Boaz.
2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008. 2008. p. 144-147 4538707 (2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008).

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

TY - GEN

T1 - The effect of GMM order and CMS on speaker recognition with reverberant speech

AU - Shabtai, Noam R.

AU - Zigel, Yaniv

AU - Rafaely, Boaz

PY - 2008/9/3

Y1 - 2008/9/3

N2 - Speaker recognition is used today in a wide range of applications. The presence of reverberation, in hands-free systems for example, results in performance degradation. The effect of reverberation on the feature vectors and its relation to optimal GMM order are investigated. Optimal model order is calculated in terms of minimum BIC and KIC, and tested for EER of a GMM-based speaker recognition system. Experimental results show that for high reverberation time, reducing model order reduces EER values of speaker recognition. The effect of CMS on state of the art GMM and AGMM-based speaker recognition systems is investigated for reverberant speech. Results show that high reverberation time reduces the effectiveness of CMS.

AB - Speaker recognition is used today in a wide range of applications. The presence of reverberation, in hands-free systems for example, results in performance degradation. The effect of reverberation on the feature vectors and its relation to optimal GMM order are investigated. Optimal model order is calculated in terms of minimum BIC and KIC, and tested for EER of a GMM-based speaker recognition system. Experimental results show that for high reverberation time, reducing model order reduces EER values of speaker recognition. The effect of CMS on state of the art GMM and AGMM-based speaker recognition systems is investigated for reverberant speech. Results show that high reverberation time reduces the effectiveness of CMS.

KW - Cepstral mean subtraction

KW - Gaussian mixture model

KW - Model order

KW - Reverberation

KW - Speaker recognition

UR - http://www.scopus.com/inward/record.url?scp=50449094185&partnerID=8YFLogxK

U2 - 10.1109/HSCMA.2008.4538707

DO - 10.1109/HSCMA.2008.4538707

M3 - Conference contribution

AN - SCOPUS:50449094185

SN - 9781424423385

T3 - 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008

SP - 144

EP - 147

BT - 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008

T2 - 2008 Hands-free Speech Communication and Microphone Arrays, HSCMA 2008

Y2 - 6 May 2008 through 8 May 2008

ER -

The effect of GMM order and CMS on speaker recognition with reverberant speech

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this