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

Attention-based encoder-decoder (AED) speech recognition model has beenwidely successful in recent years. However, the joint optimization of acousticmodel and language model in end-to-end manner has created challenges for textadaptation. In particular, effectively, quickly and inexpensively adapting texthas become a primary concern for deploying AED systems in industry. To addressthis issue, we propose a novel model, the hybrid attention-basedencoder-decoder (HAED) speech recognition model that preserves the modularityof conventional hybrid automatic speech recognition systems. Our HAED modelseparates the acoustic and language models, allowing for the use ofconventional text-based language model adaptation techniques. We demonstratethat the proposed HAED model yields 21\% Word Error Rate (WER) improvements inrelative when out-of-domain text data is used for language model adaptation,and with only a minor degradation in WER on a general test set compared withconventional AED model.

Hybrid Attention-based Encoder-decoder Model for Efficient Language Model Adaptation