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

Voice cloning technologies have found applications in a variety of areasranging from personalized speech interfaces to advertisement, robotics, and soon. Existing voice cloning systems are capable of learning speakercharacteristics and use trained models to synthesize a person's voice from onlya few audio samples. Advances in cloned speech generation technologies arecapable of generating perceptually indistinguishable speech from a bona-fidespeech. These advances pose new security and privacy threats to voice-driveninterfaces and speech-based access control systems. The state-of-the-art speechsynthesis technologies use trained or tuned generative models for cloned speechgeneration. Trained generative models rely on linear operations, learnedweights, and excitation source for cloned speech synthesis. These systems leavecharacteristic artifacts in the synthesized speech. Higher-order spectralanalysis is used to capture differentiating attributes between bona-fide andcloned audios. Specifically, quadrature phase coupling (QPC) in the estimatedbicoherence, Gaussianity test statistics, and linearity test statistics areused to capture generative model artifacts. Performance of the proposed methodis evaluated on cloned audios generated using speaker adaptation- and speakerencoding-based approaches. Experimental results for a dataset consisting of 126cloned speech and 8 bona-fide speech samples indicate that the proposed methodis capable of detecting bona-fide and cloned audios with close to a perfectdetection rate.

Securing Voice-driven Interfaces against Fake (Cloned) Audio Attacks