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

Recently, the pre-trained Transformer models have received a rising interestin the field of speech processing thanks to their great success in variousdownstream tasks. However, most fine-tuning approaches update all theparameters of the pre-trained model, which becomes prohibitive as the modelsize grows and sometimes results in overfitting on small datasets. In thispaper, we conduct a comprehensive analysis of applying parameter-efficienttransfer learning (PETL) methods to reduce the required learnable parametersfor adapting to speaker verification tasks. Specifically, during thefine-tuning process, the pre-trained models are frozen, and only lightweightmodules inserted in each Transformer block are trainable (a method known asadapters). Moreover, to boost the performance in a cross-language low-resourcescenario, the Transformer model is further tuned on a large intermediatedataset before directly fine-tuning it on a small dataset. With updating fewerthan 4% of parameters, (our proposed) PETL-based methods achieve comparableperformances with full fine-tuning methods (Vox1-O: 0.55%, Vox1-E: 0.82%,Vox1-H:1.73%).

Parameter-efficient transfer learning of pre-trained Transformer models for speaker verification using adapters