Universal Adversarial Perturbations for Speech Recognition Systems

Paarth Neekhara, Shehzeen Hussain, Prakhar Pandey, Shlomo Dubnov, Julian McAuley, Farinaz Koushanfar

In this work, we demonstrate the existence of universal adversarial audio perturbations that cause mis-transcription of audio signals by automatic speech recognition (ASR) systems. We propose an algorithm to find a single quasi-imperceptible perturbation, which when added to any arbitrary speech signal, will most likely fool the victim speech recognition model. Our experiments demonstrate the application of our proposed technique by crafting audio-agnostic universal perturbations for the state-of-the-art ASR system — Mozilla DeepSpeech. Additionally, we show that such perturbations generalize to a significant extent across models that are not available during training, by performing a transferability test on a WaveNet based ASR system.

 DOI: 10.21437/Interspeech.2019-1353

Cite as: Neekhara, P., Hussain, S., Pandey, P., Dubnov, S., McAuley, J., Koushanfar, F. (2019) Universal Adversarial Perturbations for Speech Recognition Systems. Proc. Interspeech 2019, 481-485, DOI: 10.21437/Interspeech.2019-1353.

  author={Paarth Neekhara and Shehzeen Hussain and Prakhar Pandey and Shlomo Dubnov and Julian McAuley and Farinaz Koushanfar},
  title={{Universal Adversarial Perturbations for Speech Recognition Systems}},
  booktitle={Proc. Interspeech 2019},