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

Recently, fully recurrent neural network (RNN) based end-to-end models havebeen proven to be effective for multi-speaker speech recognition in both thesingle-channel and multi-channel scenarios. In this work, we explore the use ofTransformer models for these tasks by focusing on two aspects. First, wereplace the RNN-based encoder-decoder in the speech recognition model with aTransformer architecture. Second, in order to use the Transformer in themasking network of the neural beamformer in the multi-channel case, we modifythe self-attention component to be restricted to a segment rather than thewhole sequence in order to reduce computation. Besides the model architectureimprovements, we also incorporate an external dereverberation preprocessing,the weighted prediction error (WPE), enabling our model to handle reverberatedsignals. Experiments on the spatialized wsj1-2mix corpus show that theTransformer-based models achieve 40.9% and 25.6% relative WER reduction, downto 12.1% and 6.4% WER, under the anechoic condition in single-channel andmulti-channel tasks, respectively, while in the reverberant case, our methodsachieve 41.5% and 13.8% relative WER reduction, down to 16.5% and 15.2% WER.

End-to-End Multi-speaker Speech Recognition with Transformer