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

End-to-end speech summarization (E2E SSum) directly summarizes input speechinto easy-to-read short sentences with a single model. This approach ispromising because it, in contrast to the conventional cascade approach, canutilize full acoustical information and mitigate to the propagation oftranscription errors. However, due to the high cost of collectingspeech-summary pairs, an E2E SSum model tends to suffer from training datascarcity and output unnatural sentences. To overcome this drawback, we proposefor the first time to integrate a pre-trained language model (LM), which ishighly capable of generating natural sentences, into the E2E SSum decoder viatransfer learning. In addition, to reduce the gap between the independentlypre-trained encoder and decoder, we also propose to transfer the baseline E2ESSum encoder instead of the commonly used automatic speech recognition encoder.Experimental results show that the proposed model outperforms baseline and dataaugmented models.

Transfer Learning from Pre-trained Language Models Improves End-to-End Speech Summarization