Robust Audio Patch Attacks Using Physical Sample Simulation and Adversarial Patch Noise Generation

Deep neural network (DNNs) based Automatic Speech Recognition (ASR) systems are known vulnerable to adversarial attacks that are maliciously implemented by adding small but powerful distortions to the original audio input. However, most existing methods that generate audio adversarial examples targe...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on multimedia 2022, Vol.24, p.4381-4393
Main Authors: Du, Xia, Pun, Chi-Man
Format: Article
Language:English
Subjects:
Adaptive algorithms
Adversarial robust attacks
Algorithms
Artificial neural networks
audio attack
Automatic speech recognition
Computer simulation
computer vision
Detectors
ensemble method
Impulse response
Noise
Noise generation
Perturbation
Perturbation methods
Robustness
Signal to noise ratio
Speech recognition
Training
Voice activity detection
Voice activity detectors
Voice recognition
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Deep neural network (DNNs) based Automatic Speech Recognition (ASR) systems are known vulnerable to adversarial attacks that are maliciously implemented by adding small but powerful distortions to the original audio input. However, most existing methods that generate audio adversarial examples targeting ASR models cannot achieve successful robust attacks against defense methods. This paper proposes a novel framework for robust audio patch attacks using Physical Sample Simulation (PSS) and Adversarial Patch Noise Generation (APNG). First, the proposed PSS simulated real-audio with selected room impulse response for training the adversarial patches. Second, the proposed APNG generates the imperceptible audio adversarial patch examples using the voice activity detector to hide the adversarial patch noise into the non-silent locations of the input audio. Furthermore, the design Sounds Pressure Level-based adaptive noise minimization algorithm helps us further reduce the perturbation during the attack. The experimental results show that our proposed method can achieve the highest attack success rates and SNRs in various cases, comparing with other state-of-the-art attacks.
ISSN:1520-9210
1941-0077
DOI:10.1109/TMM.2021.3116426