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 this paper, we present ECAPA2, a novel hybrid neural network architectureand training strategy to produce robust speaker embeddings. Most speakerverification models are based on either the 1D- or 2D-convolutional operation,often manifested as Time Delay Neural Networks or ResNets, respectively. Hybridmodels are relatively unexplored without an intuitive explanation whatconstitutes best practices in regard to its architectural choices. We motivatethe proposed ECAPA2 model in this paper with an analysis of current speakerverification architectures. In addition, we propose a training strategy whichmakes the speaker embeddings more robust against overlapping speech and shortutterance lengths. The presented ECAPA2 architecture and training strategyattains state-of-the-art performance on the VoxCeleb1 test sets withsignificantly less parameters than current models. Finally, we make apre-trained model publicly available to promote research on downstream tasks.

ECAPA2: A Hybrid Neural Network Architecture and Training Strategy for Robust Speaker Embeddings