English

Robustifying automatic speech recognition by extracting slowly varying features

Audio and Speech Processing 2024-11-07 v3 Machine Learning Sound Machine Learning

Abstract

In the past few years, it has been shown that deep learning systems are highly vulnerable under attacks with adversarial examples. Neural-network-based automatic speech recognition (ASR) systems are no exception. Targeted and untargeted attacks can modify an audio input signal in such a way that humans still recognise the same words, while ASR systems are steered to predict a different transcription. In this paper, we propose a defense mechanism against targeted adversarial attacks consisting in removing fast-changing features from the audio signals, either by applying slow feature analysis, a low-pass filter, or both, before feeding the input to the ASR system. We perform an empirical analysis of hybrid ASR models trained on data pre-processed in such a way. While the resulting models perform quite well on benign data, they are significantly more robust against targeted adversarial attacks: Our final, proposed model shows a performance on clean data similar to the baseline model, while being more than four times more robust.

Keywords

Cite

@article{arxiv.2112.07400,
  title  = {Robustifying automatic speech recognition by extracting slowly varying features},
  author = {Matías Pizarro and Dorothea Kolossa and Asja Fischer},
  journal= {arXiv preprint arXiv:2112.07400},
  year   = {2024}
}
R2 v1 2026-06-24T08:16:47.053Z