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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Most End-to-End SLU methods depend on the pretrained ASR or language modelfeatures for intent prediction. However, other essential information in speech,such as prosody, is often ignored. Recent research has shown improved resultsin classifying dialogue acts by incorporating prosodic information. The marginsof improvement in these methods are minimal as the neural models ignoreprosodic features. In this work, we propose prosody-attention, which uses theprosodic features differently to generate attention maps across time frames ofthe utterance. Then we propose prosody-distillation to explicitly learn theprosodic information in the acoustic encoder rather than concatenating theimplicit prosodic features. Both the proposed methods improve the baselineresults, and the prosody-distillation method gives an intent classificationaccuracy improvement of 8\% and 2\% on SLURP and STOP datasets over the prosodybaseline.

Improving End-to-End SLU performance with Prosodic Attention and Distillation