Back to results
US 2025/0028579 A1Published application
Virtual ai assistant for virtual meetings
AssigneeGoogle Inc.
InventorVictor Dantas Mehmeri
Life of this patent
Estimating the term…
Abstract
A UI is provided for presentation on each client device participating in a virtual meeting. The UI includes visual items each corresponding to one or more participants. One of the visual items includes an avatar representing a virtual assistant that has a role of one of the participants. A communication of a first participant is provided via the UI during the virtual meeting. The communication indicates a request to perform an operation. A response of the virtual assistant to the indicated request is provided via the UI during the virtual meeting. The response of the virtual assistant is at least one of a textual response or an audio response.
Claims31
A method comprising:
- providing, for presentation on each of a plurality of client devices participating in a virtual meeting, a user interface (UI) during the virtual meeting, wherein the UI comprises a plurality of visual items each corresponding to one of a plurality of participants, wherein one of the visual items comprises an avatar representing a virtual assistant that has a role of one of the plurality of participants;
- providing, via the UI during the virtual meeting, a communication of a first participant of the plurality of participants, wherein the communication of the first participant indicates a request to perform an operation;
- and
- providing, via the UI during the virtual meeting, a response of the virtual assistant to the indicated request, wherein the response of the virtual assistant is at least one of a textual response or an audio response.
1 more element
A method, comprising:
- generating, during a virtual meeting, one or more media transcriptions of a plurality of media streams generated by one or more client devices associated with a plurality of participants of a virtual meeting, wherein a media transcription of the one or more media transcriptions comprises a communication of a participant of the plurality of participants of the virtual meeting;
- based on the one or more media transcriptions and during the virtual meeting, determining that the communication of the participant of the plurality of participants indicates a request to perform an operation;
- providing, during the virtual meeting, the communication indicating the request to perform an operation as input to a generative AI model;
- obtaining an output of the generative AI model;
- and
- providing a response to the indicated request during the virtual meeting using the output of the generative AI model.
3 more elements
A method comprising:
- obtaining a meeting transcript comprising one or more media transcriptions of a plurality of media streams generated by one or more client devices associated with a plurality of participants of a virtual meeting;
- providing, as input to a generative artificial intelligence (AI) model trained to perform a plurality of operations pertaining to a virtual meeting, at least a portion of the meeting transcript and a machine-generated prompt, wherein the machine-generated prompt comprises a request to perform at least one of the plurality of operations pertaining to the virtual meeting;
- and
- obtaining an output of the generative AI model, the output comprising a result of at least one performed operation from the plurality of operations pertaining to the virtual meeting.
1 more element
A method comprising:
- receiving, from a client device of a user, a communication indicating a request to perform an operation relating to a virtual meeting;
- obtaining a meeting transcript comprising a plurality of media transcriptions generated from a plurality of media streams provided by one or more client devices associated with a plurality of participants of the virtual meeting;
- providing the communication indicative of a request and at least a portion of the meeting transcript comprising a plurality of media transcriptions as input to a generative artificial intelligence (AI) model, wherein the generative AI model is trained to perform a plurality of operations pertaining to the virtual meeting, at least one of the plurality of operations corresponding to the request indicated by the communication;
- obtaining an output of the generative AI model;
- and
- providing a response to the indicated request using the output of the generative AI model.
3 more elements
Dependencies are inferred from the claim text.