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

We present dual-attention neural biasing, an architecture designed to boostWake Words (WW) recognition and improve inference time latency on speechrecognition tasks. This architecture enables a dynamic switch for its runtimecompute paths by exploiting WW spotting to select which branch of its attentionnetworks to execute for an input audio frame. With this approach, weeffectively improve WW spotting accuracy while saving runtime compute cost asdefined by floating point operations (FLOPs). Using an in-house de-identifieddataset, we demonstrate that the proposed dual-attention network can reduce thecompute cost by $90\%$ for WW audio frames, with only $1\%$ increase in thenumber of parameters. This architecture improves WW F1 score by $16\%$ relativeand improves generic rare word error rate by $3\%$ relative compared to thebaselines.

Dual-Attention Neural Transducers for Efficient Wake Word Spotting in Speech Recognition