A multi-sample single-source model using spectrographic features for biometric authentication

Salina Abdul Samad, Dzati Athiar Ramli, Aini Hussain

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

In this paper we propose a novel approach by using spectrographic features and correlation filters as classifiers to perform speaker authentication. Visual displays (spectrograms) from speech signals produced from different persons are used as features for the verification task. In order to achieve better verification results, the exclusion of low energies and the inclusion morphological image processing steps are applied to the spectrograms. It is discovered that, by applying these two techniques, the verification performance improves significantly. For the classification modeling, Unconstrained Minimum Average Correlation Energy (UMACE) filter is implemented. We propose a multi-sample approach by fusing multiple samples from different utterances at the score level. By using the average operator, both the theoretical and empirical results show that by integrating as many samples as possible can improve the overall reliability of the system. This model is called as multi-sample single-source (MSSS) model. A digit database has been used for performance evaluation, yielding an overall performance of 99.6%.

Original languageEnglish
Title of host publication2007 6th International Conference on Information, Communications and Signal Processing, ICICS
DOIs
Publication statusPublished - 2007
Event2007 6th International Conference on Information, Communications and Signal Processing, ICICS - Singapore
Duration: 10 Dec 200713 Dec 2007

Other

Other2007 6th International Conference on Information, Communications and Signal Processing, ICICS
CitySingapore
Period10/12/0713/12/07

Fingerprint

Biometrics
Authentication
Image processing
Classifiers
Display devices

Keywords

  • Correlation filter
  • Morphological image processing
  • Multi-sample approach
  • Spectrographic features

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing

Cite this

Abdul Samad, S., Ramli, D. A., & Hussain, A. (2007). A multi-sample single-source model using spectrographic features for biometric authentication. In 2007 6th International Conference on Information, Communications and Signal Processing, ICICS [4449710] https://doi.org/10.1109/ICICS.2007.4449710

A multi-sample single-source model using spectrographic features for biometric authentication. / Abdul Samad, Salina; Ramli, Dzati Athiar; Hussain, Aini.

2007 6th International Conference on Information, Communications and Signal Processing, ICICS. 2007. 4449710.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abdul Samad, S, Ramli, DA & Hussain, A 2007, A multi-sample single-source model using spectrographic features for biometric authentication. in 2007 6th International Conference on Information, Communications and Signal Processing, ICICS., 4449710, 2007 6th International Conference on Information, Communications and Signal Processing, ICICS, Singapore, 10/12/07. https://doi.org/10.1109/ICICS.2007.4449710
Abdul Samad S, Ramli DA, Hussain A. A multi-sample single-source model using spectrographic features for biometric authentication. In 2007 6th International Conference on Information, Communications and Signal Processing, ICICS. 2007. 4449710 https://doi.org/10.1109/ICICS.2007.4449710
Abdul Samad, Salina ; Ramli, Dzati Athiar ; Hussain, Aini. / A multi-sample single-source model using spectrographic features for biometric authentication. 2007 6th International Conference on Information, Communications and Signal Processing, ICICS. 2007.
@inproceedings{1b2facf8d4c64343b393fc4e74d61447,
title = "A multi-sample single-source model using spectrographic features for biometric authentication",
abstract = "In this paper we propose a novel approach by using spectrographic features and correlation filters as classifiers to perform speaker authentication. Visual displays (spectrograms) from speech signals produced from different persons are used as features for the verification task. In order to achieve better verification results, the exclusion of low energies and the inclusion morphological image processing steps are applied to the spectrograms. It is discovered that, by applying these two techniques, the verification performance improves significantly. For the classification modeling, Unconstrained Minimum Average Correlation Energy (UMACE) filter is implemented. We propose a multi-sample approach by fusing multiple samples from different utterances at the score level. By using the average operator, both the theoretical and empirical results show that by integrating as many samples as possible can improve the overall reliability of the system. This model is called as multi-sample single-source (MSSS) model. A digit database has been used for performance evaluation, yielding an overall performance of 99.6{\%}.",
keywords = "Correlation filter, Morphological image processing, Multi-sample approach, Spectrographic features",
author = "{Abdul Samad}, Salina and Ramli, {Dzati Athiar} and Aini Hussain",
year = "2007",
doi = "10.1109/ICICS.2007.4449710",
language = "English",
isbn = "1424409837",
booktitle = "2007 6th International Conference on Information, Communications and Signal Processing, ICICS",

}

TY - GEN

T1 - A multi-sample single-source model using spectrographic features for biometric authentication

AU - Abdul Samad, Salina

AU - Ramli, Dzati Athiar

AU - Hussain, Aini

PY - 2007

Y1 - 2007

N2 - In this paper we propose a novel approach by using spectrographic features and correlation filters as classifiers to perform speaker authentication. Visual displays (spectrograms) from speech signals produced from different persons are used as features for the verification task. In order to achieve better verification results, the exclusion of low energies and the inclusion morphological image processing steps are applied to the spectrograms. It is discovered that, by applying these two techniques, the verification performance improves significantly. For the classification modeling, Unconstrained Minimum Average Correlation Energy (UMACE) filter is implemented. We propose a multi-sample approach by fusing multiple samples from different utterances at the score level. By using the average operator, both the theoretical and empirical results show that by integrating as many samples as possible can improve the overall reliability of the system. This model is called as multi-sample single-source (MSSS) model. A digit database has been used for performance evaluation, yielding an overall performance of 99.6%.

AB - In this paper we propose a novel approach by using spectrographic features and correlation filters as classifiers to perform speaker authentication. Visual displays (spectrograms) from speech signals produced from different persons are used as features for the verification task. In order to achieve better verification results, the exclusion of low energies and the inclusion morphological image processing steps are applied to the spectrograms. It is discovered that, by applying these two techniques, the verification performance improves significantly. For the classification modeling, Unconstrained Minimum Average Correlation Energy (UMACE) filter is implemented. We propose a multi-sample approach by fusing multiple samples from different utterances at the score level. By using the average operator, both the theoretical and empirical results show that by integrating as many samples as possible can improve the overall reliability of the system. This model is called as multi-sample single-source (MSSS) model. A digit database has been used for performance evaluation, yielding an overall performance of 99.6%.

KW - Correlation filter

KW - Morphological image processing

KW - Multi-sample approach

KW - Spectrographic features

UR - http://www.scopus.com/inward/record.url?scp=50449111200&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=50449111200&partnerID=8YFLogxK

U2 - 10.1109/ICICS.2007.4449710

DO - 10.1109/ICICS.2007.4449710

M3 - Conference contribution

SN - 1424409837

SN - 9781424409839

BT - 2007 6th International Conference on Information, Communications and Signal Processing, ICICS

ER -