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

End-to-End speech-to-speech translation (S2ST) is generally evaluated withtext-based metrics. This means that generated speech has to be automaticallytranscribed, making the evaluation dependent on the availability and quality ofautomatic speech recognition (ASR) systems. In this paper, we propose atext-free evaluation metric for end-to-end S2ST, named BLASER, to avoid thedependency on ASR systems. BLASER leverages a multilingual multimodal encoderto directly encode the speech segments for source input, translation output andreference into a shared embedding space and computes a score of the translationquality that can be used as a proxy to human evaluation. To evaluate ourapproach, we construct training and evaluation sets from more than 40k humanannotations covering seven language directions. The best results of BLASER areachieved by training with supervision from human rating scores. We show thatwhen evaluated at the sentence level, BLASER correlates significantly betterwith human judgment compared to ASR-dependent metrics including ASR-SENTBLEU inall translation directions and ASR-COMET in five of them. Our analysis showscombining speech and text as inputs to BLASER does not increase the correlationwith human scores, but best correlations are achieved when using speech, whichmotivates the goal of our research. Moreover, we show that using ASR forreferences is detrimental for text-based metrics.

BLASER: A Text-Free Speech-to-Speech Translation Evaluation Metric