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

Receiver operating characteristic (ROC) and detection error tradeoff (DET)curves are two widely used evaluation metrics for speaker verification. Theyare equivalent since the latter can be obtained by transforming the former'strue positive y-axis to false negative y-axis and then re-scaling both axes bya probit operator. Real-world speaker verification systems, however, usuallywork on part of the ROC curve instead of the entire ROC curve given anapplication. Therefore, we propose in this paper to use the area under part ofthe ROC curve (pAUC) as a more efficient evaluation metric for speakerverification. A Mahalanobis distance metric learning based back-end is appliedto optimize pAUC, where the Mahalanobis distance metric learning guaranteesthat the optimization objective of the back-end is a convex one so that theglobal optimum solution is achievable. To improve the performance of thestate-of-the-art speaker verification systems by the proposed back-end, wefurther propose two feature preprocessing techniques based onlength-normalization and probabilistic linear discriminant analysisrespectively. We evaluate the proposed systems on the major languages of NISTSRE16 and the core tasks of SITW. Experimental results show that the proposedback-end outperforms the state-of-the-art speaker verification back-ends interms of seven evaluation metrics.

Speaker Verification By Partial AUC Optimization With Mahalanobis Distance Metric Learning