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

Automatic dubbing (AD) is among the machine translation (MT) use cases wheretranslations should match a given length to allow for synchronicity betweensource and target speech. For neural MT, generating translations of lengthclose to the source length (e.g. within +-10% in character count), whilepreserving quality is a challenging task. Controlling MT output length comes ata cost to translation quality, which is usually mitigated with a two stepapproach of generating N-best hypotheses and then re-ranking based on lengthand quality. This work introduces a self-learning approach that allows atransformer model to directly learn to generate outputs that closely match thesource length, in short Isometric MT. In particular, our approach does notrequire to generate multiple hypotheses nor any auxiliary ranking function. Wereport results on four language pairs (English - French, Italian, German,Spanish) with a publicly available benchmark. Automatic and manual evaluationsshow that our method for Isometric MT outperforms more complex approachesproposed in the literature.

Isometric MT: Neural Machine Translation for Automatic Dubbing