Conference Call Polling Question Detection to Reduce Discussion Delays

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing conference call platforms face inefficiencies due to participants interrupting the natural flow and increasing discussion length when posing polling questions, leading to increased resource utilization and latency.

Innovation Solution

A machine learning model is trained to recognize polling questions from verbal phrases during a conference call, allowing automatic designation and display of questions without participant interruption, and populate answers based on initial responses, reducing discussion length and resource usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If participants manually pose polling questions during a conference call, then the polling function is achieved, but the natural flow of discussion is interrupted and discussion length increases

Engineering Contradiction:
Improvepolling function availabilityVSAvoiddiscussion length
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The system automatically detects polling questions from participant speech without requiring manual intervention. The speech-to-text conversion and machine learning model autonomously identify polling questions, extract answer options, and present them to participants for confirmation, eliminating the need for manual polling setup and preserving discussion flow.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system performs preliminary processing of speech to text conversion and polling question identification before the actual polling occurs. By pre-processing the discussion content and preparing polling questions in advance, the system reduces the time required during the live discussion and minimizes interruptions.

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If participants manually pose polling questions during a conference call, then the polling function is achieved, but resource utilization increases

Engineering Contradiction:
Improvepolling function availabilityVSAvoidresource utilization
Core Design Contradiction:
Ease of operationVSUse of energy by moving object

Solution Approach 1:

The system autonomously handles speech-to-text conversion, polling question detection, answer option extraction, and UI generation without requiring manual participant input. This automated approach reduces the computational overhead associated with manual polling operations and optimizes resource utilization by processing only relevant segments of the discussion.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical operations (participants manually typing or selecting polling options) with automated speech processing and machine learning-based detection. This substitution reduces the resource burden on participant devices and shifts processing to the server-side conference platform.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If participants manually pose polling questions during a conference call, then the polling function is achieved, but latency increases

Engineering Contradiction:
Improvepolling function availabilityVSAvoidresponse latency
Core Design Contradiction:
Ease of operationVSSpeed

Solution Approach 1:

The system continuously performs speech-to-text conversion and analyzes discussion content in real-time to pre-identify potential polling questions before they are fully articulated. This preliminary detection and preparation of polling questions and answer options significantly reduces the latency between question presentation and participant response.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system provides real-time feedback to participants about detected polling questions and extracted answer options, allowing for quick confirmation or correction. This feedback mechanism enables rapid iteration and reduces the overall time required for the polling process by immediately addressing any misunderstandings or errors.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4297030B1Polling questions for a conference call discussion
Publication Date: 2025.11.19 GOOGLE LLC
  • EP4297030B1 patent drawingFigure 1
  • EP4297030B1 patent drawingFigure 2
  • EP4297030B1 patent drawingFigure 3A

AI summary

Systems and methods for recognizing polling questions from a conference call discussion are provided. One or more text strings including a textual form of one or more verbal phrases provided by one or more participants of a conference call are identified. The one or more text strings are provided as input to a trained machine learning model. One or more outputs from the trained machine learning model are obtained. A level of confidence that a verbal phrase of the one or more verbal phrase provided by the one or more participants includes a question associated with polling during the conference call is extracted from the one or more outputs. In response to a determination that the level of confidence satisfies a confidence criterion, the verbal phrase is designated as a polling question presented during the conference call.