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 paper describes the University of Sydney& JD's joint submission of theIWSLT 2021 low resource speech translation task. We participated in theSwahili-English direction and got the best scareBLEU (25.3) score among all theparticipants. Our constrained system is based on a pipeline framework, i.e. ASRand NMT. We trained our models with the officially provided ASR and MTdatasets. The ASR system is based on the open-sourced tool Kaldi and this workmainly explores how to make the most of the NMT models. To reduce thepunctuation errors generated by the ASR model, we employ our previous workSlotRefine to train a punctuation correction model. To achieve bettertranslation performance, we explored the most recent effective strategies,including back translation, knowledge distillation, multi-feature reranking andtransductive finetuning. For model structure, we tried auto-regressive andnon-autoregressive models, respectively. In addition, we proposed two novelpre-train approaches, i.e. \textit{de-noising training} and\textit{bidirectional training} to fully exploit the data. Extensiveexperiments show that adding the above techniques consistently improves theBLEU scores, and the final submission system outperforms the baseline(Transformer ensemble model trained with the original parallel data) byapproximately 10.8 BLEU score, achieving the SOTA performance.

The USYD-JD Speech Translation System for IWSLT 2021