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

Masked language model (MLM) has been widely used for understanding tasks,e.g. BERT. Recently, MLM has also been used for generation tasks. The mostpopular one in speech is using Mask-CTC for non-autoregressive speechrecognition. In this paper, we take one step further, and explore thepossibility of using MLM as a non-autoregressive spell correction (SC) modelfor transformer-transducer (TT), denoted as MLM-SC. Our initial experimentsshow that MLM-SC provides no improvements on Librispeech data. The problemmight be the choice of modeling units (word pieces) and the inaccuracy of theTT confidence scores for English data. To solve the problem, we propose a masksample decoding (MS-decode) method where the masked tokens can have the choiceof being masked or not to compensate for the inaccuracy. As a result, we reducethe WER of a streaming TT from 7.6% to 6.5% on the Librispeech test-other dataand the CER from 7.3% to 6.1% on the Aishell test data, respectively.

Acoustic-aware Non-autoregressive Spell Correction with Mask Sample Decoding