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

Summarization of speech is a difficult problem due to the spontaneity of theflow, disfluencies, and other issues that are not usually encountered inwritten texts. Our work presents the first application of the BERTSum model toconversational language. We generate abstractive summaries of narratedinstructional videos across a wide variety of topics, from gardening andcooking to software configuration and sports. In order to enrich thevocabulary, we use transfer learning and pretrain the model on a few largecross-domain datasets in both written and spoken English. We also dopreprocessing of transcripts to restore sentence segmentation and punctuationin the output of an ASR system. The results are evaluated with ROUGE andContent-F1 scoring for the How2 and WikiHow datasets. We engage human judges toscore a set of summaries randomly selected from a dataset curated fromHowTo100M and YouTube. Based on blind evaluation, we achieve a level of textualfluency and utility close to that of summaries written by human contentcreators. The model beats current SOTA when applied to WikiHow articles thatvary widely in style and topic, while showing no performance regression on thecanonical CNN/DailyMail dataset. Due to the high generalizability of the modelacross different styles and domains, it has great potential to improveaccessibility and discoverability of internet content. We envision thisintegrated as a feature in intelligent virtual assistants, enabling them tosummarize both written and spoken instructional content upon request.

Abstractive Summarization of Spoken and Written Instructions with BERT