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

Despite speaker verification has achieved significant performance improvementwith the development of deep neural networks, domain mismatch is still achallenging problem in this field. In this study, we propose a novel frameworkto disentangle speaker-related and domain-specific features and apply domainadaptation on the speaker-related feature space solely. Instead of performingdomain adaptation directly on the feature space where domain information is notremoved, using disentanglement can efficiently boost adaptation performance. Tobe specific, our model's input speech from the source and target domains isfirst encoded into different latent feature spaces. The adversarial domainadaptation is conducted on the shared speaker-related feature space toencourage the property of domain-invariance. Further, we minimize the mutualinformation between speaker-related and domain-specific features for bothdomains to enforce the disentanglement. Experimental results on the VOiCESdataset demonstrate that our proposed framework can effectively generate morespeaker-discriminative and domain-invariant speaker representations with arelative 20.3% reduction of EER compared to the original ResNet-based system.

DEAAN: Disentangled Embedding and Adversarial Adaptation Network for Robust Speaker Representation Learning