English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Current adversarial attacks against speaker recognition systems (SRSs)require either white-box access or heavy black-box queries to the target SRS,thus still falling behind practical attacks against proprietary commercial APIsand voice-controlled devices. To fill this gap, we propose QFA2SR, an effectiveand imperceptible query-free black-box attack, by leveraging thetransferability of adversarial voices. To improve transferability, we presentthree novel methods, tailored loss functions, SRS ensemble, and time-freqcorrosion. The first one tailors loss functions to different attack scenarios.The latter two augment surrogate SRSs in two different ways. SRS ensemblecombines diverse surrogate SRSs with new strategies, amenable to the uniquescoring characteristics of SRSs. Time-freq corrosion augments surrogate SRSs byincorporating well-designed time-/frequency-domain modification functions,which simulate and approximate the decision boundary of the target SRS anddistortions introduced during over-the-air attacks. QFA2SR boosts the targetedtransferability by 20.9%-70.7% on four popular commercial APIs (MicrosoftAzure, iFlytek, Jingdong, and TalentedSoft), significantly outperformingexisting attacks in query-free setting, with negligible effect on theimperceptibility. QFA2SR is also highly effective when launched over the airagainst three wide-spread voice assistants (Google Assistant, Apple Siri, andTMall Genie) with 60%, 46%, and 70% targeted transferability, respectively.

QFA2SR: Query-Free Adversarial Transfer Attacks to Speaker Recognition Systems