English

Adversarial Black-Box Attacks on Automatic Speech Recognition Systems using Multi-Objective Evolutionary Optimization

Cryptography and Security 2019-07-04 v2 Machine Learning Neural and Evolutionary Computing

Abstract

Fooling deep neural networks with adversarial input have exposed a significant vulnerability in the current state-of-the-art systems in multiple domains. Both black-box and white-box approaches have been used to either replicate the model itself or to craft examples which cause the model to fail. In this work, we propose a framework which uses multi-objective evolutionary optimization to perform both targeted and un-targeted black-box attacks on Automatic Speech Recognition (ASR) systems. We apply this framework on two ASR systems: Deepspeech and Kaldi-ASR, which increases the Word Error Rates (WER) of these systems by upto 980%, indicating the potency of our approach. During both un-targeted and targeted attacks, the adversarial samples maintain a high acoustic similarity of 0.98 and 0.97 with the original audio.

Keywords

Cite

@article{arxiv.1811.01312,
  title  = {Adversarial Black-Box Attacks on Automatic Speech Recognition Systems using Multi-Objective Evolutionary Optimization},
  author = {Shreya Khare and Rahul Aralikatte and Senthil Mani},
  journal= {arXiv preprint arXiv:1811.01312},
  year   = {2019}
}

Comments

Published in Interspeech 2019

R2 v1 2026-06-23T05:03:20.418Z