Detecting Audio Attacks on ASR Systems with Dropout Uncertainty

Tejas Jayashankar, Jonathan Le Roux, Pierre Moulin


Various adversarial audio attacks have recently been developed to fool automatic speech recognition (ASR) systems. We here propose a defense against such attacks based on the uncertainty introduced by dropout in neural networks. We show that our defense is able to detect attacks created through optimized perturbations and frequency masking on a state-of-the-art end-to-end ASR system. Furthermore, the defense can be made robust against attacks that are immune to noise reduction. We test our defense on Mozilla’s CommonVoice dataset, the UrbanSound dataset, and an excerpt of the LibriSpeech dataset, showing that it achieves high detection accuracy in a wide range of scenarios.


 DOI: 10.21437/Interspeech.2020-1846

Cite as: Jayashankar, T., Roux, J.L., Moulin, P. (2020) Detecting Audio Attacks on ASR Systems with Dropout Uncertainty. Proc. Interspeech 2020, 4671-4675, DOI: 10.21437/Interspeech.2020-1846.


@inproceedings{Jayashankar2020,
  author={Tejas Jayashankar and Jonathan Le Roux and Pierre Moulin},
  title={{Detecting Audio Attacks on ASR Systems with Dropout Uncertainty}},
  year=2020,
  booktitle={Proc. Interspeech 2020},
  pages={4671--4675},
  doi={10.21437/Interspeech.2020-1846},
  url={http://dx.doi.org/10.21437/Interspeech.2020-1846}
}