Vulnerabilities of Audio-Based Biometric Authentication Systems Against Deepfake Speech Synthesis

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hong, Mengze, Jiang, Di, Xie, Zeying, Zhao, Weiwei, Wang, Guan, Zhang, Chen Jason
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911356558508032
author Hong, Mengze
Jiang, Di
Xie, Zeying
Zhao, Weiwei
Wang, Guan
Zhang, Chen Jason
author_facet Hong, Mengze
Jiang, Di
Xie, Zeying
Zhao, Weiwei
Wang, Guan
Zhang, Chen Jason
contents As audio deepfakes transition from research artifacts to widely available commercial tools, robust biometric authentication faces pressing security threats in high-stakes industries. This paper presents a systematic empirical evaluation of state-of-the-art speaker authentication systems based on a large-scale speech synthesis dataset, revealing two major security vulnerabilities: 1) modern voice cloning models trained on very small samples can easily bypass commercial speaker verification systems; and 2) anti-spoofing detectors struggle to generalize across different methods of audio synthesis, leading to a significant gap between in-domain performance and real-world robustness. These findings call for a reconsideration of security measures and stress the need for architectural innovations, adaptive defenses, and the transition towards multi-factor authentication.
format Preprint
id arxiv_https___arxiv_org_abs_2601_02914
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Vulnerabilities of Audio-Based Biometric Authentication Systems Against Deepfake Speech Synthesis
Hong, Mengze
Jiang, Di
Xie, Zeying
Zhao, Weiwei
Wang, Guan
Zhang, Chen Jason
Sound
Cryptography and Security
As audio deepfakes transition from research artifacts to widely available commercial tools, robust biometric authentication faces pressing security threats in high-stakes industries. This paper presents a systematic empirical evaluation of state-of-the-art speaker authentication systems based on a large-scale speech synthesis dataset, revealing two major security vulnerabilities: 1) modern voice cloning models trained on very small samples can easily bypass commercial speaker verification systems; and 2) anti-spoofing detectors struggle to generalize across different methods of audio synthesis, leading to a significant gap between in-domain performance and real-world robustness. These findings call for a reconsideration of security measures and stress the need for architectural innovations, adaptive defenses, and the transition towards multi-factor authentication.
title Vulnerabilities of Audio-Based Biometric Authentication Systems Against Deepfake Speech Synthesis
topic Sound
Cryptography and Security
url https://arxiv.org/abs/2601.02914