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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this study, we propose a timbre-reserved adversarial attack approach forspeaker identification (SID) to not only exploit the weakness of the SID modelbut also preserve the timbre of the target speaker in a black-box attacksetting. Particularly, we generate timbre-reserved fake audio by adding anadversarial constraint during the training of the voice conversion model. Then,we leverage a pseudo-Siamese network architecture to learn from the black-boxSID model constraining both intrinsic similarity and structural similaritysimultaneously. The intrinsic similarity loss is to learn an intrinsicinvariance, while the structural similarity loss is to ensure that thesubstitute SID model shares a similar decision boundary to the fixed black-boxSID model. The substitute model can be used as a proxy to generatetimbre-reserved fake audio for attacking. Experimental results on the AudioDeepfake Detection (ADD) challenge dataset indicate that the attack successrate of our proposed approach yields up to 60.58% and 55.38% in the white-boxand black-box scenarios, respectively, and can deceive both human beings andmachines.

Pseudo-Siamese Network based Timbre-reserved Black-box Adversarial Attack in Speaker Identification