Enabling Fast and Universal Audio Adversarial Attack Using Generative Model

TL;DR

This paper introduces fast and universal audio adversarial attack methods using generative models, significantly improving attack speed and effectiveness against DNN-based audio systems in real-time scenarios.

Contribution

The paper proposes FAPG for rapid adversarial perturbation generation and UAPG for crafting universal perturbations, advancing practical attack capabilities.

Findings

FAPG achieves up to 167X speedup over existing methods.

UAPG produces more effective universal adversarial perturbations.

Both methods outperform state-of-the-art in speed and attack success rate.

Abstract

Recently, the vulnerability of DNN-based audio systems to adversarial attacks has obtained the increasing attention. However, the existing audio adversarial attacks allow the adversary to possess the entire user's audio input as well as granting sufficient time budget to generate the adversarial perturbations. These idealized assumptions, however, makes the existing audio adversarial attacks mostly impossible to be launched in a timely fashion in practice (e.g., playing unnoticeable adversarial perturbations along with user's streaming input). To overcome these limitations, in this paper we propose fast audio adversarial perturbation generator (FAPG), which uses generative model to generate adversarial perturbations for the audio input in a single forward pass, thereby drastically improving the perturbation generation speed. Built on the top of FAPG, we further propose universal audio adversarial perturbation generator (UAPG), a scheme crafting universal adversarial perturbation that can be imposed on arbitrary benign audio input to cause misclassification. Extensive experiments show that our proposed FAPG can achieve up to 167X speedup over the state-of-the-art audio adversarial attack methods. Also our proposed UAPG can generate universal adversarial perturbation that achieves much better attack performance than the state-of-the-art solutions.