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

The Automated Speech Recognition (ASR) community experiences a major turningpoint with the rise of the fully-neural (End-to-End, E2E) approaches. At thesame time, the conventional hybrid model remains the standard choice for thepractical usage of ASR. According to previous studies, the adoption of E2E ASRin real-world applications was hindered by two main limitations: their abilityto generalize on unseen domains and their high operational cost. In this paper,we investigate both above-mentioned drawbacks by performing a comprehensivemulti-domain benchmark of several contemporary E2E models and a hybridbaseline. Our experiments demonstrate that E2E models are viable alternativesfor the hybrid approach, and even outperform the baseline both in accuracy andin operational efficiency. As a result, our study shows that the generalizationand complexity issues are no longer the major obstacle for industrialintegration, and draws the community's attention to other potential limitationsof the E2E approaches in some specific use-cases.

Are E2E ASR models ready for an industrial usage?