Speaker identification evaluation based on the speech biometric and i-vector model using the TIMIT and NTIMIT databases

Musab T.S. Al-Kaltakchi; Wai L. Woo; Satnam S. Dlay; Jonathon A. Chambers

doi:10.1109/IWBF.2017.7935102

Speaker identification evaluation based on the speech biometric and i-vector model using the TIMIT and NTIMIT databases

Musab T.S. Al-Kaltakchi, Wai L. Woo, Satnam S. Dlay, Jonathon A. Chambers

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

9 Citations (Scopus)

Abstract

Physiological and behavioural human characteristics are exploited in biometrics and performance metrics are used to measure some characteristic of an individual. The measure might lead to a one-to-one match, which is called authentication or one-from-N, and a match represents identification. In this paper, we exploit a speech biometric I-vector with low and fixed dimension of 100 to identify speakers. The main structure of the system consists of an I-vector with three fusion methods. It has low complexity and is efficient due to using an Extreme Learning Machine (ELM) classifier. The system is evaluated with 120 speakers from dialect regions one and four from both the TIMIT and NTIMIT databases in order to provide a fair comparison with our previous study based on the traditional Gaussian Mixture Model-Universal Background Model (GMM-UBM) with a Maximum Likelihood (ML) classifier system. The system shows identification rate improvement compared with the classical GMM-UBM.

Original language	English
Title of host publication	Proceedings - 2017 5th International Workshop on Biometrics and Forensics, IWBF 2017
Publisher	IEEE
ISBN (Electronic)	9781509057917
ISBN (Print)	978-1-5090-5792-4
DOIs	https://doi.org/10.1109/IWBF.2017.7935102
Publication status	Published - 29 May 2017
Externally published	Yes
Event	5th International Workshop on Biometrics and Forensics, IWBF 2017 - Coventry, United Kingdom Duration: 4 Apr 2017 → 5 Apr 2017

Conference

Conference	5th International Workshop on Biometrics and Forensics, IWBF 2017
Country/Territory	United Kingdom
City	Coventry
Period	4/04/17 → 5/04/17

Keywords

Fusion approaches
Speaker identification
Speech biometric and I-vector
TIMIT and NTIMIT speech corpora

Access to Document

10.1109/IWBF.2017.7935102

Cite this

@inproceedings{2dc4b589889e49e1b15e84759cdb91d7,

title = "Speaker identification evaluation based on the speech biometric and i-vector model using the TIMIT and NTIMIT databases",

abstract = "Physiological and behavioural human characteristics are exploited in biometrics and performance metrics are used to measure some characteristic of an individual. The measure might lead to a one-to-one match, which is called authentication or one-from-N, and a match represents identification. In this paper, we exploit a speech biometric I-vector with low and fixed dimension of 100 to identify speakers. The main structure of the system consists of an I-vector with three fusion methods. It has low complexity and is efficient due to using an Extreme Learning Machine (ELM) classifier. The system is evaluated with 120 speakers from dialect regions one and four from both the TIMIT and NTIMIT databases in order to provide a fair comparison with our previous study based on the traditional Gaussian Mixture Model-Universal Background Model (GMM-UBM) with a Maximum Likelihood (ML) classifier system. The system shows identification rate improvement compared with the classical GMM-UBM.",

keywords = "Fusion approaches, Speaker identification, Speech biometric and I-vector, TIMIT and NTIMIT speech corpora",

author = "Al-Kaltakchi, {Musab T.S.} and Woo, {Wai L.} and Dlay, {Satnam S.} and Chambers, {Jonathon A.}",

year = "2017",

month = may,

day = "29",

doi = "10.1109/IWBF.2017.7935102",

language = "English",

isbn = "978-1-5090-5792-4",

booktitle = "Proceedings - 2017 5th International Workshop on Biometrics and Forensics, IWBF 2017",

publisher = "IEEE",

address = "United States",

note = "5th International Workshop on Biometrics and Forensics, IWBF 2017 ; Conference date: 04-04-2017 Through 05-04-2017",

}

Al-Kaltakchi, MTS, Woo, WL, Dlay, SS & Chambers, JA 2017, Speaker identification evaluation based on the speech biometric and i-vector model using the TIMIT and NTIMIT databases. in Proceedings - 2017 5th International Workshop on Biometrics and Forensics, IWBF 2017. IEEE, 5th International Workshop on Biometrics and Forensics, IWBF 2017, Coventry, United Kingdom, 4/04/17. https://doi.org/10.1109/IWBF.2017.7935102

Speaker identification evaluation based on the speech biometric and i-vector model using the TIMIT and NTIMIT databases. / Al-Kaltakchi, Musab T.S.; Woo, Wai L.; Dlay, Satnam S. et al.
Proceedings - 2017 5th International Workshop on Biometrics and Forensics, IWBF 2017. IEEE, 2017.

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

TY - GEN

T1 - Speaker identification evaluation based on the speech biometric and i-vector model using the TIMIT and NTIMIT databases

AU - Al-Kaltakchi, Musab T.S.

AU - Woo, Wai L.

AU - Dlay, Satnam S.

AU - Chambers, Jonathon A.

PY - 2017/5/29

Y1 - 2017/5/29

N2 - Physiological and behavioural human characteristics are exploited in biometrics and performance metrics are used to measure some characteristic of an individual. The measure might lead to a one-to-one match, which is called authentication or one-from-N, and a match represents identification. In this paper, we exploit a speech biometric I-vector with low and fixed dimension of 100 to identify speakers. The main structure of the system consists of an I-vector with three fusion methods. It has low complexity and is efficient due to using an Extreme Learning Machine (ELM) classifier. The system is evaluated with 120 speakers from dialect regions one and four from both the TIMIT and NTIMIT databases in order to provide a fair comparison with our previous study based on the traditional Gaussian Mixture Model-Universal Background Model (GMM-UBM) with a Maximum Likelihood (ML) classifier system. The system shows identification rate improvement compared with the classical GMM-UBM.

AB - Physiological and behavioural human characteristics are exploited in biometrics and performance metrics are used to measure some characteristic of an individual. The measure might lead to a one-to-one match, which is called authentication or one-from-N, and a match represents identification. In this paper, we exploit a speech biometric I-vector with low and fixed dimension of 100 to identify speakers. The main structure of the system consists of an I-vector with three fusion methods. It has low complexity and is efficient due to using an Extreme Learning Machine (ELM) classifier. The system is evaluated with 120 speakers from dialect regions one and four from both the TIMIT and NTIMIT databases in order to provide a fair comparison with our previous study based on the traditional Gaussian Mixture Model-Universal Background Model (GMM-UBM) with a Maximum Likelihood (ML) classifier system. The system shows identification rate improvement compared with the classical GMM-UBM.

KW - Fusion approaches

KW - Speaker identification

KW - Speech biometric and I-vector

KW - TIMIT and NTIMIT speech corpora

U2 - 10.1109/IWBF.2017.7935102

DO - 10.1109/IWBF.2017.7935102

M3 - Conference contribution

AN - SCOPUS:85021839349

SN - 978-1-5090-5792-4

BT - Proceedings - 2017 5th International Workshop on Biometrics and Forensics, IWBF 2017

PB - IEEE

T2 - 5th International Workshop on Biometrics and Forensics, IWBF 2017

Y2 - 4 April 2017 through 5 April 2017

ER -

Speaker identification evaluation based on the speech biometric and i-vector model using the TIMIT and NTIMIT databases

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this