US-20260128919-A1 - CONFERENCING SYSTEM WITH MULTI-MODAL SENSING AND CONTEXTUAL MODEL TRAINING

US20260128919A1US 20260128919 A1US20260128919 A1US 20260128919A1US-20260128919-A1

Abstract

A conferencing system may position a sensor array in a meeting space and detect, with the sensor array, an activity of a participant present in the meeting space. A computing device connected to the sensor array may assign at least one identifier to describe the activity prior to characterizing the at least one identifier as a behavioral context. In response to the behavioral context, the computing device may alter at least one operational parameter of the sensor array.

Inventors

James Michael Dallas
Matthew Skogmo
Damian Andrea FRICK
Ryan Pring

Assignees

QSC, LLC

Dates

Publication Date: 20260507
Application Date: 20251104

Claims (20)

1 . A method comprising: positioning a sensor array in a meeting space; detecting, with the sensor array, an activity of a participant present in the meeting space; assigning, with a computing device connected to the sensor array, at least one identifier to describe the activity; characterizing, with the computing device, the at least one identifier as a behavioral context; and altering, with the computing device, at least one operational parameter of the sensor array in response to the behavioral context.
2 . The method of claim 1 , wherein the behavioral context is characterized by the computing device as an emotional state of the participant.
3 . The method of claim 1 , wherein the at least one identifier describes an emotion to the detected activity.
4 . The method of claim 3 , wherein the emotion is sarcasm and the detected activity is listening.
5 . The method of claim 1 , wherein the sensor array comprises at least one acoustic sensor, at least one optical sensor, and at least one environmental sensor.
6 . The method of claim 1 , wherein the computing device utilizes a context strategy to assign the at least one identifier to describe the activity.
7 . The method of claim 6 , wherein the context strategy proactively sets rules and policies for operation of the sensor array that aid in the efficient characterization of behavior of the participant into the at least one identifier.
8 . The method of claim 1 , wherein at least one optical sensor and at least one acoustic sensor are used to detect the activity of the participant.
9 . The method of claim 1 , wherein the sensor array is configured as a single component positioned in a central location of the meeting space.
10 . The method of claim 1 , wherein the computing device formats the at least one identifier to train an intelligence model.
11 . A method comprising: positioning a sensor array in a meeting space; sensing, with the sensor array, at least one condition of the meeting space; detecting, with the sensor array, a first participant present in the meeting space and a second participant present in the meeting space; detecting, with the sensor array, an activity of the first participant; assigning, with a computing device connected to the sensor array, at least one identifier to describe the activity; characterizing, with the computing device, the at least one identifier as a relationship based on a relationship strategy; and verifying, with the computing device, the relationship in response to detected behavior of the first participant.
12 . The method of claim 11 , wherein the computing device assigns at least one behavioral identifier to the first participant after verification of the relationship between the first participant and the second participant.
13 . The method of claim 11 , wherein the computing device is configured to assign at least one identifier to the second participant based on activity of the second participant detected by the sensor array, the at least one identifier assigned to the second participant being different than the at least one identifier assigned to the first participant.
14 . The method of claim 11 , wherein the computing device assigns an initial relationship between the first participant and the second participant prior to detecting the activity of the first participant.
15 . The method of claim 11 , wherein the relationship strategy alters operational parameters of the sensor array to assign multiple different identifiers to the activity.
16 . The method of claim 11 , wherein the computing device trains an intelligence model with the at least one identifier.
17 . The method of claim 11 , wherein an intelligence model is used by the computing device to characterize the at least one identifier as a behavioral context.
18 . The method of claim 11 , further comprising identifying, with a computing device connected to the sensor array, a first profile corresponding to the first participant and a second profile corresponding with the second participant; and generating, with the computing device, a relationship strategy based on the at least one condition, the first profile, and the second profile.
19 . The method of claim 11 , wherein the computing device is configured to recharacterize the relationship in response to the detected behavior of the first participant.
20 . A system comprising: a sensor array positioned within a meeting space, the sensor array having an initial set of operating parameters; a computing device connected to the sensor array; wherein the sensor array is configured to detect activity of at least two participants; wherein the computing device is configured to assign at least one identifier to the detected activity; wherein the computing device is configured to generate a relationship strategy prescribing a relationship set of operating parameters for the sensor array to detect interpersonal relationships between meeting participants; wherein the computing device is configured to designate an initial relationship status to a pair of meeting participants in response to the activation of the relationship set of operating parameters for the sensor array; wherein the computing device is configured to generate a context strategy prescribing a context set of operating parameters for the sensor array to detect behavior of at least two meeting participants; wherein the computing device is configured to assign identifiers to detected behavior of a meeting participant with the computing device, the identifiers describing meaning corresponding with the detected behavior; wherein the computing device is configured to generate a conferencing strategy prescribing a content set of operating parameters for the sensor array to collect audio content and video content from meeting participants with customized accuracy; wherein the computing device is configured to conduct context analysis on the at least one identifier to determine a real-time participant status based on an intelligence model accessed by the computing device; wherein the computing device is configured to choose an intelligence model to train with the at least one identifier; and wherein the computing device is configured to format the at least one identifier to train the intelligence model.

Description

CROSS REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Application Nos. 63/716,521 filed November 5, 2024, which is currently pending, the disclosure of which is hereby incorporated by reference herein in its entirety. SUMMARY Embodiments of the present disclosure are generally directed to a conferencing system employing multi-modal sensing to intelligently understand a conferencing environment that can be utilized to gather conference participant behavior, assign context to the gathered information, and train one or more intelligence models with the participant’s contextual actions. In some embodiments of a conferencing system, a sensor array may be positioned in a meeting space and used to detect activity of a participant present in the meeting space. A computing device connected to the sensor array may assign at least one identifier to describe the activity prior to characterizing the at least one identifier as a behavioral context. In response to the behavioral context, the computing device may alter at least one operational parameter of the sensor array. Embodiments of a conferencing system may position a sensor array in a meeting space and sense, with the sensor array, at least one condition of the meeting space. The sensor array may detect a first participant present in the meeting space, a second participant present in the meeting space, and an activity of the first participant. A computing device connected to the sensor array may assign at least one identifier to describe the activity and characterize the at least one identifier as a relationship based on a relationship strategy. The computing device may verify the assigned relationship in response to detected behavior of the first participant. Other embodiments of a conferencing system may position a sensor array within a meeting space with the sensor array having an initial set of operating parameters. A computing device may be connected to the sensor array and activity of at least two participants may be detected by the sensor array. The computing device may assign at least one identifier to the detected activity and generate a relationship strategy prescribing a relationship set of operating parameters for the sensor array to detect interpersonal relationships between meeting participants. The computing device may designate an initial relationship status to a pair of meeting participants in response to the activation of the relationship set of operating parameters for the sensor array. The computing may generate a context strategy prescribing a context set of operating parameters for the sensor array to detect behavior of at least two meeting participants and assign identifiers to detected behavior of a meeting participant with the identifiers describing meaning corresponding with the detected behavior. The computing device may generate a conferencing strategy prescribing a content set of operating parameters for the sensor array to collect audio content and video content from meeting participants with customized accuracy and conduct context analysis on the at least one identifier to determine a real-time participant status based on an intelligence model accessed by the computing device. The computing device may choose an intelligence model to train with the at least one identifier and format the at least one identifier to train the intelligence model. A conferencing system, in accordance with some embodiments, may have a sensor array positioned in a meeting space. An initial set of operating parameters is installed for the sensor array before detecting characteristics of the meeting space with the sensor array. Meeting participants in the meeting space are identified with the sensor array and a relationship strategy is generated with a computing device connected to the sensor array with the relationship strategy prescribing a relationship set of operating parameters for the sensor array to detect interpersonal relationships between meeting participants. Next, the computing device designates an initial relationship status to a pair of meeting participants in response to the activation of the relationship set of operating parameters for the sensor array. A context strategy is generated with the computing device that prescribes a context set of operating parameters for the sensor array to detect behavior of at least one meeting participant. The computing device may then assign identifiers to detected behavior of a meeting participant that describe meaning corresponding with the detected behavior. A conferencing strategy is generated with the computing device that prescribes a content set of operating parameters for the sensor array to collect audio content and video content from meeting participants with customized accuracy. The computing device then formats the identifiers to train at least one intelligence model. Other embodiments of a conferencing system may position a sensor array in a meeting space wi