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/

Zendy Plus

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

Zendy Tools

Zendy Open

Deep learning is at the core of recent spoken language understanding (SLU)related tasks. More precisely, deep neural networks (DNNs) drasticallyincreased the performances of SLU systems, and numerous architectures have beenproposed. In the real-life context of theme identification of telephoneconversations, it is common to hold both a human, manual (TRS) and anautomatically transcribed (ASR) versions of the conversations. Nonetheless, anddue to production constraints, only the ASR transcripts are considered to buildautomatic classifiers. TRS transcripts are only used to measure theperformances of ASR systems. Moreover, the recent performances in term ofclassification accuracy, obtained by DNN related systems are close to theperformances reached by humans, and it becomes difficult to further increasethe performances by only considering the ASR transcripts. This paper proposesto distillates the TRS knowledge available during the training phase within theASR representation, by using a new generative adversarial network calledM2H-GAN to generate a TRS-like version of an ASR document, to improve the themeidentification performances.

M2H-GAN: A GAN-based Mapping from Machine to Human Transcripts for Speech Understanding