跳到主要导航 跳到搜索 跳到主要内容

Inaudible adversarial perturbations for targeted attack in speaker recognition

  • Northwestern Polytechnical University Xian

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

35 引用 (Scopus)

摘要

Speaker recognition is a popular topic in biometric authentication and many deep learning approaches have achieved extraordinary performances. However, it has been shown in both image and speech applications that deep neural networks are vulnerable to adversarial examples. In this study, we aim to exploit this weakness to perform targeted adversarial attacks against the x-vector based speaker recognition system. We propose to generate inaudible adversarial perturbations based on the psychoacoustic principle of frequency masking, achieving targeted white-box attacks to speaker recognition system. Specifically, we constrict the perturbation under the masking threshold of original audio, instead of using a common lp norm to measure the perturbations. Experiments on Aishell-1 corpus show that our approach yields up to 98.5% attack success rate to arbitrary gender speaker targets, while retaining indistinguishable attribute to listeners. Furthermore, we also achieve an effective speaker attack when applying the proposed approach to a completely irrelevant waveform, such as music.

源语言英语
主期刊名Interspeech 2020
出版商International Speech Communication Association
页4228-4232
页数5
ISBN(印刷版)9781713820697
DOI
出版状态已出版 - 2020
活动21st Annual Conference of the International Speech Communication Association, INTERSPEECH 2020 - Shanghai, 中国
期限: 25 10月 2020 → 29 10月 2020

丛书

姓名Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
卷2020-October
ISSN(印刷版)2308-457X
ISSN(电子版)1990-9772

会议

会议21st Annual Conference of the International Speech Communication Association, INTERSPEECH 2020
国家/地区中国
市Shanghai
时期25/10/20 → 29/10/20

学术指纹

探究 'Inaudible adversarial perturbations for targeted attack in speaker recognition' 的科研主题。它们共同构成独一无二的学术指纹。

引用此