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

Self-attention networks (SAN) have been introduced into automatic speechrecognition (ASR) and achieved state-of-the-art performance owing to itssuperior ability in capturing long term dependency. One of the key ingredientsis the self-attention mechanism which can be effectively performed on the wholeutterance level. In this paper, we try to investigate whether even moreinformation beyond the whole utterance level can be exploited and beneficial.We propose to apply self-attention layer with augmented memory to ASR.Specifically, we first propose a variant model architecture which combines deepfeed-forward sequential memory network (DFSMN) with self-attention layers toform a better baseline model compared with a purely self-attention network.Then, we propose and compare two kinds of additional memory structures addedinto self-attention layers. Experiments on large-scale LVCSR tasks show that onfour individual test sets, the DFSMN-SAN architecture outperforms vanilla SANencoder by 5% relatively in character error rate (CER). More importantly, theadditional memory structure provides further 5% to 11% relative improvement inCER.

DFSMN-SAN with Persistent Memory Model for Automatic Speech Recognition