Secure speech template protection in speaker verification system

Beng Jin Teoh, Lee Ying Chong

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Due to biometric template characteristics that are susceptible to non-revocable and privacy invasion, cancellable biometrics has been introduced to tackle these issues. In this paper, we present a two-factor cancellable formulation for speech biometrics, which we refer as probabilistic random projection (PRP). PRP offers strong protection on speech template by hiding the actual speech feature through the random subspace projection process. Besides, the speech template is replaceable and can be reissued when it is compromised. Our proposed method enables the generation of different speech templates from the same speech feature, which means linkability is not exited between the speech templates. The formulation of the cancellable biometrics retains its performance as for the conventional biometric. Besides that, we also propose 2D subspace projection techniques for speech feature extraction, namely 2D Principle Component Analysis (2DPCA) and 2D CLAss-Featuring Information Compression (2DCLAFIC) to accommodate the requirements of PRP formulation.

Original languageEnglish
Pages (from-to)150-163
Number of pages14
JournalSpeech Communication
Volume52
Issue number2
DOIs
Publication statusPublished - 2010 Feb 1

Fingerprint

Speaker Verification
Template
Biometrics
projection
Random Projection
Formulation
Subspace
Projection
Principle Component Analysis
Speech
Invasion
invasion
Feature Extraction
Privacy
privacy
Feature extraction
Compression
biometrics
Requirements

All Science Journal Classification (ASJC) codes

  • Software
  • Modelling and Simulation
  • Communication
  • Language and Linguistics
  • Linguistics and Language
  • Computer Vision and Pattern Recognition
  • Computer Science Applications

Cite this

@article{097522511b214b35809a097c115928c4,
title = "Secure speech template protection in speaker verification system",
abstract = "Due to biometric template characteristics that are susceptible to non-revocable and privacy invasion, cancellable biometrics has been introduced to tackle these issues. In this paper, we present a two-factor cancellable formulation for speech biometrics, which we refer as probabilistic random projection (PRP). PRP offers strong protection on speech template by hiding the actual speech feature through the random subspace projection process. Besides, the speech template is replaceable and can be reissued when it is compromised. Our proposed method enables the generation of different speech templates from the same speech feature, which means linkability is not exited between the speech templates. The formulation of the cancellable biometrics retains its performance as for the conventional biometric. Besides that, we also propose 2D subspace projection techniques for speech feature extraction, namely 2D Principle Component Analysis (2DPCA) and 2D CLAss-Featuring Information Compression (2DCLAFIC) to accommodate the requirements of PRP formulation.",
author = "Teoh, {Beng Jin} and Chong, {Lee Ying}",
year = "2010",
month = "2",
day = "1",
doi = "10.1016/j.specom.2009.09.003",
language = "English",
volume = "52",
pages = "150--163",
journal = "Speech Communication",
issn = "0167-6393",
publisher = "Elsevier",
number = "2",

}

Secure speech template protection in speaker verification system. / Teoh, Beng Jin; Chong, Lee Ying.

In: Speech Communication, Vol. 52, No. 2, 01.02.2010, p. 150-163.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Secure speech template protection in speaker verification system

AU - Teoh, Beng Jin

AU - Chong, Lee Ying

PY - 2010/2/1

Y1 - 2010/2/1

N2 - Due to biometric template characteristics that are susceptible to non-revocable and privacy invasion, cancellable biometrics has been introduced to tackle these issues. In this paper, we present a two-factor cancellable formulation for speech biometrics, which we refer as probabilistic random projection (PRP). PRP offers strong protection on speech template by hiding the actual speech feature through the random subspace projection process. Besides, the speech template is replaceable and can be reissued when it is compromised. Our proposed method enables the generation of different speech templates from the same speech feature, which means linkability is not exited between the speech templates. The formulation of the cancellable biometrics retains its performance as for the conventional biometric. Besides that, we also propose 2D subspace projection techniques for speech feature extraction, namely 2D Principle Component Analysis (2DPCA) and 2D CLAss-Featuring Information Compression (2DCLAFIC) to accommodate the requirements of PRP formulation.

AB - Due to biometric template characteristics that are susceptible to non-revocable and privacy invasion, cancellable biometrics has been introduced to tackle these issues. In this paper, we present a two-factor cancellable formulation for speech biometrics, which we refer as probabilistic random projection (PRP). PRP offers strong protection on speech template by hiding the actual speech feature through the random subspace projection process. Besides, the speech template is replaceable and can be reissued when it is compromised. Our proposed method enables the generation of different speech templates from the same speech feature, which means linkability is not exited between the speech templates. The formulation of the cancellable biometrics retains its performance as for the conventional biometric. Besides that, we also propose 2D subspace projection techniques for speech feature extraction, namely 2D Principle Component Analysis (2DPCA) and 2D CLAss-Featuring Information Compression (2DCLAFIC) to accommodate the requirements of PRP formulation.

UR - http://www.scopus.com/inward/record.url?scp=70449095873&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449095873&partnerID=8YFLogxK

U2 - 10.1016/j.specom.2009.09.003

DO - 10.1016/j.specom.2009.09.003

M3 - Article

AN - SCOPUS:70449095873

VL - 52

SP - 150

EP - 163

JO - Speech Communication

JF - Speech Communication

SN - 0167-6393

IS - 2

ER -