Adversarial Attack and Defense for Commercial Black-box Chinese-English Speech Recognition Systems
Open Access
- 11 December 2024
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Privacy and Security
- Vol. 28 (1) , 1-27
Abstract
The attacker can generate adversarial examples (AEs) to stealthily mislead automatic speech recognition (ASR) models, raising significant concerns about the security of intelligent voice control (IVC) devices. Existing adversarial attacks mainly generate AEs to mislead ASR models to output specific target English commands (e.g., open the door). However, it remains unknown whether AEs can be used to issue commands in other languages to attack commercial black-box ASR models. In this article, taking Chinese phrases (e.g., 支付宝付款) and “Chinese–English code-switching” phrases (e.g., 关闭GPS) as the target commands, we propose adversarial attacks for commercial multilingual ASR models. In particular, if a multilingual speech recognition model can recognize Chinese and English, we call it a Chinese–English speech recognition model. In English, the meaning of “支付宝付款” and “关闭GPS” are “Alipay payment” and “turn off GPS”, respectively. In detail, we generate transferable AEs based on the open-sourced conventional DataTang Mandarin ASR model. Given 55 target commands, the success rate for generating AEs of them is up to 96% and 80% for Aliyun ASR API and Tencentyun ASR API, respectively. Our AEs can trigger actual attack actions on voice assistants (e.g., Apple Siri, Xiaomi Xiaoaitongxue) or spread malicious messages through ASR API services, while the target commands in the AEs are inaudible to human beings. 1 Finally, by analyzing the spectrum differences between benign audio clips and AEs, we propose a general defense against adversarial audio attacks.Keywords
Funding Information
- National Natural Science Foundation of China (62202064)
- NSFC (92270204)
- CAS Project for Young Scientists in Basic Research (YSBR-118)
- Youth Innovation Promotion Association CAS
This publication has 18 references indexed in Scilit:
- PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection2023
- Towards the universal defense for query-based audio adversarial attacks on speech recognition systemCybersecurity, 2023
- A Survey on Voice Assistant Security: Attacks and CountermeasuresACM Computing Surveys, 2022
- SPECPATCH2022
- SoK: A Modularized Approach to Study the Security of Automatic Speech Recognition SystemsACM Transactions on Privacy and Security, 2022
- Black-box Adversarial Attacks on Commercial Speech Platforms with Minimal Information2021
- HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden UnitsIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021
- Imperio: Robust Over-the-Air Adversarial Examples for Automatic Speech Recognition Systems2020
- SirenAttack: Generating Adversarial Audio for End-to-End Acoustic Systems2020
- Audio Adversarial Examples: Targeted Attacks on Speech-to-Text2018