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 emergence of self-supervised representation (i.e., wav2vec 2.0) allowsspeaker-recognition approaches to process spoken signals through foundationmodels built on speech data. Nevertheless, effective fusion on therepresentation requires further investigating, due to the inclusion of fixed orsub-optimal temporal pooling strategies. Despite of improved strategiesconsidering graph learning and graph attention factors, non-injectiveaggregation still exists in the approaches, which may influence the performancefor speaker recognition. In this regard, we propose a speaker recognitionapproach using Isomorphic Graph ATtention network (IsoGAT) on self-supervisedrepresentation. The proposed approach contains three modules of representationlearning, graph attention, and aggregation, jointly considering learning on theself-supervised representation and the IsoGAT. Then, we perform experiments forspeaker recognition tasks on VoxCeleb1\&2 datasets, with the correspondingexperimental results demonstrating the recognition performance for the proposedapproach, compared with existing pooling approaches on the self-supervisedrepresentation.

Speaker Recognition Using Isomorphic Graph Attention Network Based Pooling on Self-Supervised Representation