Generating customized meeting insights based on user interactions and meeting media

A non-transitory computer readable storage medium comprising instructions that, when executed by at least one processor, cause a computer system to:

• receive media data associated with a meeting, the media data comprising audio or video;
• analyze the media data and one or more user inputs detected by one or more client devices during the meeting to determine one or more relevant portions of the meeting for a user associated with the meeting, the one or more relevant portions including information related to an action item associated with a user, and the one or more user inputs comprising an interaction by a user with a keyboard input, a mouse input, a touch screen input, or a microphone input indicating a specific portion of the audio or video of the media data according to one or more timestamps of the one or more user inputs and one or more corresponding times in the audio or video of the media data;
• generate, utilizing a machine-learning model processing the media data and the one or more user inputs in response to determining the one or more relevant portions of the meeting, content comprising information associated with the action item indicating a task assigned to a user associated with the meeting to perform by a date or a time based on the specific portion of the audio or video of the media data and the one or more user inputs;
• and
• provide the content comprising the action item to a client device associated with the user by providing an interactive indication of the action item within a meeting summary.

A system comprising:

• at least one processor;
• and
• a non-transitory computer memory comprising instructions that, when executed by the at least one processor, cause the system to:
• receive first media data from one or more client devices, the first media data comprising audio or video associated with a first meeting;
• analyze the first media data and one or more user inputs detected by the one or more client devices to determine a relevant portion of the first meeting for a user associated with the first meeting, the one or more user inputs comprising an interaction by a user with a keyboard input, a mouse input, a touch screen input, or a microphone input indicating a specific portion of the audio or video of the first media data according to one or more timestamps of the one or more user inputs and one or more corresponding times in the audio or video of the first media data;
• generate, from the audio or video of the first media data and the one or more user inputs in response to determining the relevant portion of the first meeting, meeting insight data comprising content including one or more action items indicating one or more tasks assigned to a user associated with the first meeting to perform by a date or a time based on the relevant portion of the audio or video of the first media data of the first meeting and the one or more user inputs;
• train a machine-learning model based on the first media data, the one or more user inputs, and the meeting insight data comprising an indication of the one or more action items within a meeting summary;
• and
• utilize the trained machine-learning model to generate content based on second media data comprising audio associated with a second meeting.