Inaudible Manipulation of Voice-Enabled Devices through BackDoor Using Robust Adversarial Audio Attacks: Invited Paper

Morriel Kasher, Michael Zhao, Aryeh Greenberg, Devin Gulati, Silvija Kokalj-Filipovic, Predrag Spasojevic

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

The BackDoor system provides a method for inaudibly transmitting messages that are recorded by unmodified receiver microphones as if they were transmitted audibly. Adversarial Audio attacks allow for an audio sample to sound like one message but be transcribed by a speech processing neural network as a different message. This study investigates the potential applications of Adversarial Audio through the BackDoor system to manipulate voice-enabled devices, or VEDs, without detection by humans or other nearby microphones. We discreetly transmit voice commands by applying robust, noise-resistant adversarial audio perturbations through BackDoor on top of a predetermined speech or music base sample to achieve a desired target transcription. Our analysis compares differing base carriers, target phrases, and perturbation strengths for maximal effectiveness through BackDoor. We determined that such an attack is feasible and that the desired adversarial properties of the audio sample are maintained even when transmitted through BackDoor.

Original languageEnglish (US)
Title of host publicationWiseML 2021 - Proceedings of the 3rd ACM Workshop on Wireless Security and Machine Learning
PublisherAssociation for Computing Machinery, Inc
Pages37-42
Number of pages6
ISBN (Electronic)9781450385619
DOIs
StatePublished - Jun 28 2021
Externally publishedYes
Event3rd ACM Workshop on Wireless Security and Machine Learning, WiseML 2021 - Virtual, Online, United Arab Emirates
Duration: Jul 2 2021 → …

Publication series

NameWiseML 2021 - Proceedings of the 3rd ACM Workshop on Wireless Security and Machine Learning

Conference

Conference3rd ACM Workshop on Wireless Security and Machine Learning, WiseML 2021
Country/TerritoryUnited Arab Emirates
CityVirtual, Online
Period7/2/21 → …

All Science Journal Classification (ASJC) codes

  • Software
  • Artificial Intelligence
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Inaudible Manipulation of Voice-Enabled Devices through BackDoor Using Robust Adversarial Audio Attacks: Invited Paper'. Together they form a unique fingerprint.

Cite this