US 11,956,091 B1
Video conference content auto-retrieval and focus based on learned relevance
Victor Carbune, Winterthur (CH); Daniel Keysers, Stallikon (CH); and Thomas Deselaers, Zurich (CH)
Assigned to Google LLC, Mountain View, CA (US)
Filed by GOOGLE LLC, Mountain View, CA (US)
Filed on Oct. 24, 2022, as Appl. No. 17/972,515.
Application 17/972,515 is a continuation of application No. 16/730,484, filed on Dec. 30, 2019, granted, now 11,483,170.
Int. Cl. G06F 15/16 (2006.01); H04L 12/18 (2006.01); H04M 3/56 (2006.01)
CPC H04L 12/1831 (2013.01) [H04L 12/1818 (2013.01); H04L 12/1822 (2013.01); H04M 3/566 (2013.01); H04M 3/568 (2013.01)] 23 Claims
OG exemplary drawing
 
1. A method comprising:
receiving a plurality of audio streams and a plurality of video streams from a plurality of client devices participating in a video conference;
determining, based on the plurality of audio streams, a first subject being discussed during the video conference at a first point in time;
determining a video stream from the plurality of video streams that is most relevant to the first subject being discussed during the video conference at the first point in time;
providing the determined video stream to the plurality of client devices for presentation in a video conference user interface (UI) on the plurality of client devices while the first subject is being discussed during the video conference;
determining, at a second point in time during the video conference, that the first subject has changed to a second subject;
identifying a document that is relevant to the second subject;
providing, for presentation on the plurality of client devices, a modified video conference UI including a suggestion pertaining to the identified document; and
upon receiving an indication of an acceptance of the suggestion, providing a video stream associated with a presentation of the identified document in the video conference UI on the plurality of client devices while the second subject is being discussed during the video conference.