Trojan Attacks and Defense for Speech Recognition

Wei Zong, Yang Wai Chow, Willy Susilo, Jongkil Kim

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Mobile devices commonly employ speech recognition (SR) techniques to facilitate user interaction. Typical voice assistants on mobile devices detect a wake word or phrase before allowing users to use voice commands. While the core functionality of contemporary SR systems relies on deep learning, researchers have shown that deep learning suffers from various security issues. Among these security threats, Trojan attacks in particular have attracted great interest in the research community. To conduct a Trojan attack, an adversary must stealthily modify a target model, such that the compromised model will output a predefined label whenever presented with a trigger. Most work in the literature has focused on Trojan attacks for image recognition, and there is limited work in the SR domain. Due to the increasing use of SR systems in daily devices, such as mobile phones, Trojan attacks for SR pose a great threat to the public and is therefore an important topic of concern to mobile internet security. Despite its growing importance, there has not been an extensive review conducted on Trojan attacks for SR. This paper fills this gap by presenting an overview of existing techniques of conducting Trojan attacks and defending against them for SR. The purpose is to provide researchers with an in-depth comparison of current methods and the challenges faced in this important research area.

Original languageEnglish
Title of host publicationMobile Internet Security - 5th International Symposium, MobiSec 2021, Revised Selected Papers
EditorsIlsun You, Hwankuk Kim, Taek-Young Youn, Francesco Palmieri, Igor Kotenko
PublisherSpringer Science and Business Media Deutschland GmbH
Pages195-210
Number of pages16
ISBN (Print)9789811695759
DOIs
StatePublished - 2022
Event5th International Symposium on Mobile Internet Security, MobiSec 2021 - Jeju Island, Korea, Republic of
Duration: 7 Oct 20219 Oct 2021

Publication series

NameCommunications in Computer and Information Science
Volume1544 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference5th International Symposium on Mobile Internet Security, MobiSec 2021
Country/TerritoryKorea, Republic of
CityJeju Island
Period7/10/219/10/21

Bibliographical note

Publisher Copyright:
© 2022, Springer Nature Singapore Pte Ltd.

Keywords

  • Deep learning
  • Machine learning
  • Speech recognition
  • Trojan attacks

Fingerprint

Dive into the research topics of 'Trojan Attacks and Defense for Speech Recognition'. Together they form a unique fingerprint.

Cite this