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

This work studies the use of attention masking in transformer transducerbased speech recognition for building a single configurable model for differentdeployment scenarios. We present a comprehensive set of experiments comparingfixed masking, where the same attention mask is applied at every frame, withchunked masking, where the attention mask for each frame is determined by chunkboundaries, in terms of recognition accuracy and latency. We then explore theuse of variable masking, where the attention masks are sampled from a targetdistribution at training time, to build models that can work in differentconfigurations. Finally, we investigate how a single configurable model can beused to perform both first pass streaming recognition and second pass acousticrescoring. Experiments show that chunked masking achieves a better accuracy vslatency trade-off compared to fixed masking, both with and without FastEmit. Wealso show that variable masking improves the accuracy by up to 8% relative inthe acoustic re-scoring scenario.

Variable Attention Masking for Configurable Transformer Transducer Speech Recognition