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

Current speaker recognition systems primarily rely on supervised approaches,constrained by the scale of labeled datasets. To boost the system performance,researchers leverage large pretrained models such as WavLM to transfer learnedhigh-level features to the downstream speaker recognition task. However, thisapproach introduces extra parameters as the pretrained model remains in theinference stage. Another group of researchers directly apply self-supervisedmethods such as DINO to speaker embedding learning, yet they have not exploredits potential on large-scale in-the-wild datasets. In this paper, we presentthe effectiveness of DINO training on the large-scale WenetSpeech dataset andits transferability in enhancing the supervised system performance on theCNCeleb dataset. Additionally, we introduce a confidence-based data filteringalgorithm to remove unreliable data from the pretraining dataset, leading tobetter performance with less training data. The associated pretrained models,confidence files, pretraining and finetuning scripts will be made available inthe Wespeaker toolkit.

Leveraging In-the-Wild Data for Effective Self-Supervised Pretraining in Speaker Recognition