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 introduces NoRefER, a novel referenceless quality metric forautomatic speech recognition (ASR) systems. Traditional reference-based metricsfor evaluating ASR systems require costly ground-truth transcripts. NoRefERovercomes this limitation by fine-tuning a multilingual language model forpair-wise ranking ASR hypotheses using contrastive learning with Siamesenetwork architecture. The self-supervised NoRefER exploits the known qualityrelationships between hypotheses from multiple compression levels of an ASR forlearning to rank intra-sample hypotheses by quality, which is essential formodel comparisons. The semi-supervised version also uses a referenced datasetto improve its inter-sample quality ranking, which is crucial for selectingpotentially erroneous samples. The results indicate that NoRefER correlateshighly with reference-based metrics and their intra-sample ranks, indicating ahigh potential for referenceless ASR evaluation or a/b testing.

NoRefER: a Referenceless Quality Metric for Automatic Speech Recognition via Semi-Supervised Language Model Fine-Tuning with Contrastive Learning