New Microsoft patent hints at a topic-based summarizer for meetings

A new Microsoft patent reveals details about a topic-based summarizer for meetings and presentations. The summarizer would be helpful for people who were unable to attend the meeting or for users to refresh on what occurred during the meeting – without the effort of watching the entire recording.

Targeted towards video conferencing, the system will generate a summary of a session among the users based on the determined relative importance of each statement made during the session to other statements made using hierarchical agglomerative clustering, a bottom-up approach of cluster analysis.

The patent, filed in February last year, has been granted to Microsoft with credits to Jeet Sunil Mody, Shalendra Chhabra, and Senthil Kumar Velayutham as the inventors.

Topic based summarizer for meetings and presentations using hierarchical agglomerative clustering

Disclosed are systems, methods, and non-transitory computer-readable media for a meeting-topic based summarizer that uses hierarchical agglomerative clustering (HAC). A meeting summarization system generates representative vectors for each statement in a text. Each statement includes one or more terms and each representative vector indicates a relative importance of its respective statement to the text based on the one or more terms included in the respective statement. The meeting summarization system generates vector clusters based on the representative vectors and determines topics of the text based on the statements represented by the representative vectors included in each vector cluster. The meeting summarization system generates a summary of the text based on the determined topics.

The obvious application of such a system will be in Microsoft Teams to properly summarize meetings and presentations, similar to the common minutes-of-the-meeting practice but done automatically.

Teams is set to get a live transcription feature next month that will allow users to view a live transcription of speech during a call (available for download post a meeting) with the content attributed to each speaker.