Adversarial command detection using parallel Speech Recognition systems

Simple item page
dc.contributor.authorCheng, Peng
dc.contributor.authorSankar, M. S. Arun
dc.contributor.authorBagci, Ibrahim Ethem
dc.contributor.authorRoedig, Utz
dc.contributor.funderScience Foundation Irelanden
dc.date.accessioned2021-09-16T10:32:15Z
dc.date.available2021-09-16T10:32:15Z
dc.date.issued2021-10
dc.description.abstractPersonal Voice Assistants (PVAs) such as Apple's Siri, Amazon's Alexa and Google Home are now commonplace. PVAs are susceptible to adversarial commands; an attacker is able to modify an audio signal such that humans do not notice this modification but the Speech Recognition (SR) will recognise a command of the attacker's choice. In this paper we describe a defence method against such adversarial commands. By using a second SR in parallel to the main SR of the PVA it is possible to detect adversarial commands. It is difficult for an attacker to craft an adversarial command that is able to force two different SR into recognising the adversarial command while ensuring inaudibility. We demonstrate the feasibility of this defence mechanism for practical setups. For instance, our evaluation shows that such system can be tuned to detect 50% of adversarial commands while not impacting on normal PVA use.en
dc.description.sponsorshipScience Foundation Ireland (SFI 19/FFP/6775)en
dc.description.statusPeer revieweden
dc.description.versionAccepted Versionen
dc.format.mimetypeapplication/pdfen
dc.identifier.citationCheng, P., Sankar M. S., A., Bagci, I. E. and Roedig,U. (2021) 'Adversarial command detection using parallel Speech Recognition systems’, ESORICS 2021, 26th European Symposium on Research in Computer Security, Lecture Notes in Computer Science, 13106, pp. 238-255. doi: 10.1007/978-3-030-95484-0_15en
dc.identifier.doi10.1007/978-3-030-95484-0_15
dc.identifier.endpage238
dc.identifier.startpage255
dc.identifier.urihttps://hdl.handle.net/10468/11931
dc.language.isoenen
dc.publisherSpringeren
dc.relation.urihttps://link.springer.com/chapter/10.1007/978-3-030-95484-0_15en
dc.rightsFor the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. Copyright Published Article: © Springer Nature Switzerland AG 2022en
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/en
dc.subjectPersonal Voice Assistantsen
dc.subjectPVAsen
dc.subjectSpeech Recognitionen
dc.subjectSRen
dc.subjectAdversarial commandsen
dc.titleAdversarial command detection using parallel Speech Recognition systemsen
dc.typeConference itemen
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Adversarial_Command_Detection_using_Parallel_SR_ADIoT.pdf
Size:
833.22 KB
Format:
Adobe Portable Document Format
Description:
Accepted Version
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
2.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Computer Science - Conference Items