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 rhythm of synthetic speech is usually too smooth, which causes that thefundamental frequency (F0) of synthetic speech is significantly different fromthat of real speech. It is expected that the F0 feature contains thediscriminative information for the fake speech detection (FSD) task. In thispaper, we propose a novel F0 subband for FSD. In addition, to effectively modelthe F0 subband so as to improve the performance of FSD, the spatialreconstructed local attention Res2Net (SR-LA Res2Net) is proposed.Specifically, Res2Net is used as a backbone network to obtain multiscaleinformation, and enhanced with a spatial reconstruction mechanism to avoidlosing important information when the channel group is constantly superimposed.In addition, local attention is designed to make the model focus on the localinformation of the F0 subband. Experimental results on the ASVspoof 2019 LAdataset show that our proposed method obtains an equal error rate (EER) of0.47% and a minimum tandem detection cost function (min t-DCF) of 0.0159,achieving the state-of-the-art performance among all of the single systems.

Spatial Reconstructed Local Attention Res2Net with F0 Subband for Fake Speech Detection