Video Conference Context Association via Summary Generation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users joining video conferences late often miss crucial contextual information, leading to difficulty in understanding the discussion and resulting in a poor user experience and wasted time.

Innovation Solution

A video conferencing application that uses speech recognition and natural language processing to generate summaries of previous discussions, displaying them alongside the live video conference, allowing late joiners to quickly catch up on the context through subtitles and summaries.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If users join video conferences late, then they can participate in ongoing discussions, but they miss crucial contextual information from earlier discussions

Engineering Contradiction:

ImproveAbility to join video conference at any timeVSAvoidMissed contextual information from early discussion

Core Design Contradiction:

Adaptability or versatilityVSLoss of information

Solution Approach 1:

The system generates and displays summaries of previous discussions before late joiners enter the meeting. These summaries are prepared in advance and made available immediately when users join, allowing them to catch up on contextual information without delaying their participation in ongoing discussions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary mechanism (summary generation and display system) that bridges the gap between early discussion content and late joiners. The system processes audio from previous discussions, generates condensed summaries, and presents them as contextual information, enabling late joiners to understand references made during the conversation.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Loss of information

If the system displays all previous discussion content, then late joiners can access complete context, but the interface becomes complex and overwhelming

Engineering Contradiction:

ImproveComprehensive contextual information availabilityVSAvoidUser interface complexity

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The system extracts only the most relevant contextual information from extensive previous discussions and presents it as condensed summaries. Rather than displaying all original content, the system identifies and extracts key points, topics, and actionable information, reducing the interface complexity while maintaining essential context for late joiners.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the discussion content into distinct summary units organized by topic or time periods. This segmentation allows the interface to present comprehensive context in an organized, digestible format rather than as a monolithic block of information, reducing cognitive load and interface complexity.

Inventive Principle:

Principle #1Segmentation

3Loss of information

If the system processes and displays continuous summaries, then context is always available, but processing time and computational resources increase

Engineering Contradiction:

ImproveReal-time contextual information availabilityVSAvoidTime for processing and displaying summaries

Core Design Contradiction:

Loss of informationVSLoss of time

Solution Approach 1:

Instead of continuously processing every word of audio in real-time, the system processes audio in periodic segments or batches. Summaries are generated at specific intervals or when new discussion topics emerge, providing contextual information periodically rather than continuously, which reduces processing time and computational resources while maintaining effective context availability.

Inventive Principle:

Principle #19Periodic action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Improves user experience by enabling late joiners to quickly understand the current discussion, enhancing productivity by providing essential context and reducing the time spent catching up on missed information.

Implementation Method 1

the video conference application can translate the first received audio (e.g., a first user speaking) into a first subtitle using speech recognition methodologies (e.g., automatic speech recognition, computer speech recognition, speech to text, etc.)

Methodology Applied

Scientific EffectAutomatic speech recognition:

Implementation Method 2

The video conference application can process the first subtitle and generate a first summary (e.g., 'Introduction,' 'Recent Updates,' etc.) using natural language process algorithms

Methodology Applied

Scientific EffectNatural language processing:

Data Source

PatentUS20240187271A1Systems and methods for associating context to subtitles during live events

Publication Date: 2024.06.06 ADEIA GUIDES INC

AI summary

Systems and methods are provided herein for providing context to users who access video conferences late. This may be accomplished by a system receiving an audio segment of a video conference and generating a subtitle corresponding to the audio segment. The system may determine a summary relating to the audio segment and then display the subtitle, summary, and video conference on a device. The system allows a user, who accesses a video conference late, to quickly and accurately understand the current video conference discussion, improving the user's experience and increasing the productivity of the video conference.