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

Audio adversarial examples (AEs) have posed significant security challengesto real-world speaker recognition systems. Most black-box attacks still requirecertain information from the speaker recognition model to be effective (e.g.,keeping probing and requiring the knowledge of similarity scores). This workaims to push the practicality of the black-box attacks by minimizing theattacker's knowledge about a target speaker recognition model. Although it isnot feasible for an attacker to succeed with completely zero knowledge, weassume that the attacker only knows a short (or a few seconds) speech sample ofa target speaker. Without any probing to gain further knowledge about thetarget model, we propose a new mechanism, called parrot training, to generateAEs against the target model. Motivated by recent advancements in voiceconversion (VC), we propose to use the one short sentence knowledge to generatemore synthetic speech samples that sound like the target speaker, called parrotspeech. Then, we use these parrot speech samples to train a parrot-trained(PT)surrogate model for the attacker. Under a joint transferability and perceptionframework, we investigate different ways to generate AEs on the PT model(called PT-AEs) to ensure the PT-AEs can be generated with high transferabilityto a black-box target model with good human perceptual quality. Real-worldexperiments show that the resultant PT-AEs achieve the attack success rates of45.8% - 80.8% against the open-source models in the digital-line scenario and47.9% - 58.3% against smart devices, including Apple HomePod (Siri), AmazonEcho, and Google Home, in the over-the-air scenario.

Parrot-Trained Adversarial Examples: Pushing the Practicality of Black-Box Audio Attacks against Speaker Recognition Models