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

The spoof speech detection (SSD) is the essential countermeasure forautomatic speaker verification systems. Although SSD with magnitude features inthe frequency domain has shown promising results, the phase information alsocan be important to capture the artefacts of certain types of spoofing attacks.Thus, both magnitude and phase features must be considered to ensure thegeneralization ability to diverse types of spoofing attacks. In this paper, weinvestigate the failure reason of feature-level fusion of the previous worksthrough the entropy analysis from which we found that the randomness differencebetween magnitude and phase features is large, which can interrupt thefeature-level fusion via backend neural network; thus, we propose a phasenetwork to reduce that difference. Our SSD system: phase network equippedRes2Net achieved significant performance improvement, specifically in thespoofing attack for which the phase information is considered to be important.Also, we demonstrate our SSD system in both known- and unknown-kind SSDscenarios for practical applications.

Phase-Aware Spoof Speech Detection Based on Res2Net with Phase Network