A comparative evaluation on data transformation approach for artificial speech detection

The rise of voice biometrics has transformed user authentication and offered enhanced security and convenience while phasing out less secure methods. Despite these advancements, Automatic Speaker Verification (ASV) systems remain vulnerable to spoofing, particularly with artificial speech generated...

Full description

Bibliographic Details
Main Authors: Choon Beng Tan, Mohd Hanafi Ahmad Hijazi
Format: Proceedings
Language:English
English
Published: EDP Sciences 2024
Subjects:
Online Access:https://eprints.ums.edu.my/id/eprint/39280/1/ABSTRACT.pdf
https://eprints.ums.edu.my/id/eprint/39280/2/FULL%20TEXT.pdf
_version_ 1825715675766718464
author Choon Beng Tan
Mohd Hanafi Ahmad Hijazi
author_facet Choon Beng Tan
Mohd Hanafi Ahmad Hijazi
author_sort Choon Beng Tan
collection UMS
description The rise of voice biometrics has transformed user authentication and offered enhanced security and convenience while phasing out less secure methods. Despite these advancements, Automatic Speaker Verification (ASV) systems remain vulnerable to spoofing, particularly with artificial speech generated swiftly using advanced speech synthesis and voice conversion algorithms. A recent data transformation technique achieved an impressive Equal Error Rate (EER) of 1.42% on the ASVspoof 2019 Logical Access Dataset. While this approach predominantly relies on Support Vector Machine (SVM) as the backend classifier for artificial speech detection, it is vital to explore a broader range of classifiers to enhance resilience. This paper addresses this research gap by systematically assessing classifier efficacy in artificial speech detection. The objectives are twofold: first, to evaluate various classifiers, not limited to SVM, and identify those best suited for artificial speech detection; second, to compare this approach's performance with existing methods. The evaluation demonstrated SVM-Polynomial as the top-performing classifier, surpassing the end-to-end learning approach. This work contributes to a deeper understanding of classifier efficacy and equips researchers and practitioners with a diversified toolkit for building robust ASV spoofing detection systems.
first_indexed 2024-09-24T00:50:37Z
format Proceedings
id ums.eprints-39280
institution Universiti Malaysia Sabah
language English
English
last_indexed 2024-09-24T00:50:37Z
publishDate 2024
publisher EDP Sciences
record_format dspace
spelling ums.eprints-392802024-07-26T02:50:44Z https://eprints.ums.edu.my/id/eprint/39280/ A comparative evaluation on data transformation approach for artificial speech detection Choon Beng Tan Mohd Hanafi Ahmad Hijazi BF231-299 Sensation. Aesthesiology QA75-76.95 Calculating machines The rise of voice biometrics has transformed user authentication and offered enhanced security and convenience while phasing out less secure methods. Despite these advancements, Automatic Speaker Verification (ASV) systems remain vulnerable to spoofing, particularly with artificial speech generated swiftly using advanced speech synthesis and voice conversion algorithms. A recent data transformation technique achieved an impressive Equal Error Rate (EER) of 1.42% on the ASVspoof 2019 Logical Access Dataset. While this approach predominantly relies on Support Vector Machine (SVM) as the backend classifier for artificial speech detection, it is vital to explore a broader range of classifiers to enhance resilience. This paper addresses this research gap by systematically assessing classifier efficacy in artificial speech detection. The objectives are twofold: first, to evaluate various classifiers, not limited to SVM, and identify those best suited for artificial speech detection; second, to compare this approach's performance with existing methods. The evaluation demonstrated SVM-Polynomial as the top-performing classifier, surpassing the end-to-end learning approach. This work contributes to a deeper understanding of classifier efficacy and equips researchers and practitioners with a diversified toolkit for building robust ASV spoofing detection systems. EDP Sciences 2024 Proceedings PeerReviewed text en https://eprints.ums.edu.my/id/eprint/39280/1/ABSTRACT.pdf text en https://eprints.ums.edu.my/id/eprint/39280/2/FULL%20TEXT.pdf Choon Beng Tan and Mohd Hanafi Ahmad Hijazi (2024) A comparative evaluation on data transformation approach for artificial speech detection. https://www.researchgate.net/publication/378178682_A_Comparative_Evaluation_on_Data_Transformation_Approach_for_Artificial_Speech_Detection
spellingShingle BF231-299 Sensation. Aesthesiology
QA75-76.95 Calculating machines
Choon Beng Tan
Mohd Hanafi Ahmad Hijazi
A comparative evaluation on data transformation approach for artificial speech detection
title A comparative evaluation on data transformation approach for artificial speech detection
title_full A comparative evaluation on data transformation approach for artificial speech detection
title_fullStr A comparative evaluation on data transformation approach for artificial speech detection
title_full_unstemmed A comparative evaluation on data transformation approach for artificial speech detection
title_short A comparative evaluation on data transformation approach for artificial speech detection
title_sort comparative evaluation on data transformation approach for artificial speech detection
topic BF231-299 Sensation. Aesthesiology
QA75-76.95 Calculating machines
url https://eprints.ums.edu.my/id/eprint/39280/1/ABSTRACT.pdf
https://eprints.ums.edu.my/id/eprint/39280/2/FULL%20TEXT.pdf
work_keys_str_mv AT choonbengtan acomparativeevaluationondatatransformationapproachforartificialspeechdetection
AT mohdhanafiahmadhijazi acomparativeevaluationondatatransformationapproachforartificialspeechdetection
AT choonbengtan comparativeevaluationondatatransformationapproachforartificialspeechdetection
AT mohdhanafiahmadhijazi comparativeevaluationondatatransformationapproachforartificialspeechdetection