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

In recent years, self-supervised learning paradigm has received extensiveattention due to its great success in various down-stream tasks. However, thefine-tuning strategies for adapting those pre-trained models to speakerverification task have yet to be fully explored. In this paper, we analyzeseveral feature extraction approaches built on top of a pre-trained model, aswell as regularization and learning rate schedule to stabilize the fine-tuningprocess and further boost performance: multi-head factorized attentive poolingis proposed to factorize the comparison of speaker representations intomultiple phonetic clusters. We regularize towards the parameters of thepre-trained model and we set different learning rates for each layer of thepre-trained model during fine-tuning. The experimental results show our methodcan significantly shorten the training time to 4 hours and achieve SOTAperformance: 0.59%, 0.79% and 1.77% EER on Vox1-O, Vox1-E and Vox1-H,respectively.

An attention-based backend allowing efficient fine-tuning of transformer models for speaker verification