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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Summaries of meetings are very important as they convey the essential contentof discussions in a concise form. Generally, it is time consuming to read andunderstand the whole documents. Therefore, summaries play an important role asthe readers are interested in only the important context of discussions. Inthis work, we address the task of meeting document summarization. Automaticsummarization systems on meeting conversations developed so far have beenprimarily extractive, resulting in unacceptable summaries that are hard toread. The extracted utterances contain disfluencies that affect the quality ofthe extractive summaries. To make summaries much more readable, we propose anapproach to generating abstractive summaries by fusing important content fromseveral utterances. We first separate meeting transcripts into various topicsegments, and then identify the important utterances in each segment using asupervised learning approach. The important utterances are then combinedtogether to generate a one-sentence summary. In the text generation step, thedependency parses of the utterances in each segment are combined together tocreate a directed graph. The most informative and well-formed sub-graphobtained by integer linear programming (ILP) is selected to generate aone-sentence summary for each topic segment. The ILP formulation reducesdisfluencies by leveraging grammatical relations that are more prominent innon-conversational style of text, and therefore generates summaries that iscomparable to human-written abstractive summaries. Experimental results showthat our method can generate more informative summaries than the baselines. Inaddition, readability assessments by human judges as well as log-likelihoodestimates obtained from the dependency parser show that our generated summariesare significantly readable and well-formed.

Generating Abstractive Summaries from Meeting Transcripts