Detecting Audio Attacks on ASR Systems with Dropout Uncertainty

2020/06/02 by Jayashankar, Tejas, Roux, Jonathan Le, Moulin, Pierre
#Audio and Speech Processing (eess.AS) #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Sound (cs.SD) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2006.01906

Abstract

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.

Related

• Catch Me If You Can: Blackbox Adversarial Attacks on Automatic Speech Recognition using Frequency Masking
• Detecting Adversarial Examples for Speech Recognition via Uncertainty Quantification
• WaveGuard: Understanding and Mitigating Audio Adversarial Examples
• Adversarial Attacks and Defenses for Speech Recognition Systems
• Perceptual-based deep-learning denoiser as a defense against adversarial\n attacks on ASR systems