Network-based communication session copilot

Patent metadata

Field	Value
Publication number	US-12475887-B2
Application number	US-202318213511-A
Country	US
Kind code	B2
Filing date	Jun 23, 2023
Priority date	Feb 27, 2023
Publication date	Nov 18, 2025
Grant date	Nov 18, 2025

Abstract

A system for providing a personalized assistant within a network-based communication service includes one or more processors and a memory storage device storing instructions thereon. During a network-based communication session, the system receives a query from a computing device of a first communication session participant and processes the query by determining that a second communication session participant has shared content via a content sharing feature of the network-based communication service. In response, the system provides the query and at least a portion of the shared content as input to a model, which dynamically constructs a prompt for use as input with a generative language model. The system then presents the prompt as input to the generative language model, receives a response as output from the generative language model, and causes presentation of the response to be presented to the communication session participant.

First claim

What is claimed is:

1 . A system for providing a personalized assistant within a network-based communication service, the system comprising:
one or more processors; and a memory storage device storing instructions thereon, which, when executed by the one or more processors, cause the system to perform operations comprising:
during a network-based communication session, receiving a query from a computing device of a first communication session participant;
processing the query by:
determining that a second communication session participant has shared content via a content sharing feature of the network-based communication service;
in response to the determining, providing the query and at least a portion of the shared content as input to a model, wherein the model processes the query and the portion of shared content to dynamically construct a prompt for use as input with a generative language model;
providing the prompt as input to the generative language model;
receiving, as output from the generative language model, a response; and
causing presentation of the response to be presented to the communication session participant.

2 . The system of claim 1 , wherein the model comprises a rules-based engine that is configured to dynamically construct the prompt with a first instruction when the content shared via the content sharing feature of the network-based communication service satisfies a first rule-based condition, and to dynamically construct the prompt with a second instruction when the content shared via the content sharing feature of the network-based communication service does not satisfy the first rule-based condition.

3 . The system of claim 1 , wherein the model comprises a rules-based engine that is configured to dynamically construct the prompt by selecting, with a first selection criteria, a first portion of content from a meeting transcript to include as context in the prompt, when the content shared via the content sharing feature of the network-based communication service satisfies a first rule-based condition, and to dynamically construct the prompt by selecting, with a second selection criteria, a second portion of content from the meeting transcript to include as context in the prompt, when the content shared via the content sharing feature of the network-based communication service does not satisfy the first rule-based condition.

4 . The system of claim 1 , wherein the prompt comprises a first portion, representing context, and a second portion, representing an instruction, and dynamically constructing the prompt further comprises:
constructing the prompt to include, as context, a portion of the content shared via the content sharing feature of the network-based communication service, by: processing a prompt template for the prompt, the prompt template comprising a content selection criteria; selecting, using the content selection criteria of the prompt template, the portion of the content shared via the content sharing feature of the network-based communication service, to include as context in the prompt.

5 . The system of claim 1 , wherein the prompt comprises a first portion, representing context, and a second portion, representing an instruction, and dynamically constructing the prompt further comprises:
constructing the prompt to include, as a part of the instruction, a portion of the content shared via the content sharing feature of the network-based communication service, by: processing a prompt template for the prompt, the prompt template comprising a content selection criteria; selecting, using the content selection criteria of the prompt template, the portion of the content shared via the content sharing feature of the network-based communication service, to include as part of the instruction in the prompt.

6 . The system of claim 1 , wherein the prompt comprises a first portion, representing context, and a second portion, representing an instruction, and dynamically constructing the prompt further comprises:
constructing the prompt to include a portion of the content shared via the content sharing feature of the network-based communication service, as context; and selecting a first instruction from a plurality of instructions, based on a result of an evaluation of the content shared via the content sharing feature of the network-based communication service.

7 . The system of claim 1 , wherein the query is a free-text query specified by the first communication session participant, and dynamically constructing the prompt further comprises:
using the content shared via the content sharing feature of the network-based communication service and the free-text query as input to the model, the model comprising a machine learning model that has been trained to generate the prompt using as input to the machine learning model the content and the free-text query.

8 . A method for providing a personalized assistant within a network-based communication service, the method comprising:
during a network-based communication session, receiving a query from a computing device of a first communication session participant; processing the query by:
determining that a second communication session participant has shared content via a content sharing feature of the network-based communication service;
in response to the determining, providing the query and at least a portion of the shared content as input to a model, wherein the model processes the query and the portion of shared content to dynamically construct a prompt for use as input with a generative language model;
providing the prompt as input to the generative language model;
receiving, as output from the generative language model, a response; and
causing presentation of the response to be presented to the communication session participant.

9 . The method of claim 8 , wherein the model comprises a rules-based engine that is configured to dynamically construct the prompt with a first instruction when the content shared via the content sharing feature of the network-based communication service satisfies a first rule-based condition, and to dynamically construct the prompt with a second instruction when the content shared via the content sharing feature of the network-based communication service does not satisfy the first rule-based condition.

10 . The method of claim 8 , wherein the model comprises a rules-based engine that is configured to dynamically construct the prompt by selecting, with a first selection criteria, a first portion of content from a meeting transcript to include as context in the prompt, when the content shared via the content sharing feature of the network-based communication service satisfies a first rule-based condition, and to dynamically construct the prompt by selecting, with a second selection criteria, a second portion of content from the meeting transcript to include as context in the prompt, when the content shared via the content sharing feature of the network-based communication service does not satisfy the first rule-based condition.

11 . The method of claim 8 , wherein the prompt comprises a first portion, representing context, and a second portion, representing an instruction, and dynamically constructing the prompt further comprises:
constructing the prompt to include, as context, a portion of the content shared via the content sharing feature of the network-based communication service, by: processing a prompt template for the prompt, the prompt template comprising a content selection criteria; selecting, using the content selection criteria of the prompt template, the portion of the content shared via the content sharing feature of the network-bas

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Lu xiao yan
• Kantor Amir
• Priness Ido
• Cupala Shiraz Jitendra
• Carter Kevin Michael
• Miller Adi
• Ranjan Kumud
• Gupta Shyam
• Jain Gautam
• Cenberoglu Yasemin
• Ifrach Shai
• Maliah Shlomi
• Teevan Jaime
• Ye Lan

Classifications

• G10L15/063

  Training · CPC title

• G10L2015/223

  Execution procedure of a spoken command · CPC title

• G10L2015/221

  Announcement of recognition results · CPC title

• G06Q10/10

  Office automation; Time management · CPC title

• G10L15/26

  Speech to text systems (G10L15/08 takes precedence) · CPC title