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-activated systems are integrated into a variety of desktop, mobile, andInternet-of-Things (IoT) devices. However, voice spoofing attacks, such asimpersonation and replay attacks, in which malicious attackers synthesize thevoice of a victim or simply replay it, have brought growing security concerns.Existing speaker verification techniques distinguish individual speakers viathe spectrographic features extracted from an audible frequency range of voicecommands. However, they often have high error rates and/or long delays. In thispaper, we explore a new direction of human voice research by scrutinizing theunique characteristics of human speech at the ultrasound frequency band. Ourresearch indicates that the high-frequency ultrasound components (e.g. speechfricatives) from 20 to 48 kHz can significantly enhance the security andaccuracy of speaker verification. We propose a speaker verification system,SUPERVOICE that uses a two-stream DNN architecture with a feature fusionmechanism to generate distinctive speaker models. To test the system, we createa speech dataset with 12 hours of audio (8,950 voice samples) from 127participants. In addition, we create a second spoofed voice dataset to evaluateits security. In order to balance between controlled recordings and real-worldapplications, the audio recordings are collected from two quiet rooms by 8different recording devices, including 7 smartphones and an ultrasoundmicrophone. Our evaluation shows that SUPERVOICE achieves 0.58% equal errorrate in the speaker verification task, it only takes 120 ms for testing anincoming utterance, outperforming all existing speaker verification systems.Moreover, within 91 ms processing time, SUPERVOICE achieves 0% equal error ratein detecting replay attacks launched by 5 different loudspeakers.

SuperVoice: Text-Independent Speaker Verification Using Ultrasound Energy in Human Speech