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

Spoken language identification (LID) technologies have improved in recentyears from discriminating largely distinct languages to discriminating highlysimilar languages or even dialects of the same language. One aspect that hasbeen mostly neglected, however, is discrimination of languages for multilingualspeakers, despite being a primary target audience of many systems that utilizeLID technologies. As we show in this work, LID systems can have a high averageaccuracy for most combinations of languages while greatly underperforming forothers when accented speech is present. We address this by usingcoarser-grained targets for the acoustic LID model and integrating its outputswith interaction context signals in a context-aware model to tailor the systemto each user. This combined system achieves an average 97% accuracy across alllanguage combinations while improving worst-case accuracy by over 60% relativeto our baseline.

Improving Language Identification for Multilingual Speakers