English

Whisper Smarter, not Harder: Adversarial Attack on Partial Suppression

Sound 2025-09-09 v2 Cryptography and Security Machine Learning Audio and Speech Processing

Abstract

Currently, Automatic Speech Recognition (ASR) models are deployed in an extensive range of applications. However, recent studies have demonstrated the possibility of adversarial attack on these models which could potentially suppress or disrupt model output. We investigate and verify the robustness of these attacks and explore if it is possible to increase their imperceptibility. We additionally find that by relaxing the optimisation objective from complete suppression to partial suppression, we can further decrease the imperceptibility of the attack. We also explore possible defences against these attacks and show a low-pass filter defence could potentially serve as an effective defence.

Keywords

Cite

@article{arxiv.2508.09994,
  title  = {Whisper Smarter, not Harder: Adversarial Attack on Partial Suppression},
  author = {Zheng Jie Wong and Bingquan Shen},
  journal= {arXiv preprint arXiv:2508.09994},
  year   = {2025}
}

Comments

14 pages, 7 figures

R2 v1 2026-07-01T04:48:31.506Z