Video Quality Switching Using Predictive Volume Thresholds

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Video conferencing systems experience a time delay when a non-speaking user transitions to a speaking user, as the lower quality video feed takes 3-4 seconds to switch to a higher quality feed, affecting user experience and resource management.

Innovation Solution

A method where a first computing device in a multiway video session sends a video feed with a pre-determined quality level to a second device, determining that a volume level reaches a threshold, and promptly switches to a higher quality level based on this determination, potentially using predictive analysis to anticipate the new speaking user.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If the system uses scaling and layering to display speakers on larger images with higher quality video feed, then the visual quality for active speakers is improved, but a time delay of 3-4 seconds occurs when switching video quality from non-speakers to speakers

Engineering Contradiction:
Improvevideo qualityVSAvoidswitching delay
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The system proactively sends higher quality video feeds to the server before the user actually speaks, based on predictive analysis of conversation patterns. This preliminary action ensures that when the user does speak, the high quality feed is already available, eliminating the 3-4 second switching delay while maintaining adaptive resource management.

Inventive Principle:
Principle #10Preliminary action

2Use of energy by moving object

If the system sends lower quality video feed to non-speaking users, then network and system resources are optimized, but user experience deteriorates when there is a delay in switching to higher quality feed upon becoming a speaker

Engineering Contradiction:
Improvenetwork resource usageVSAvoiduser experience
Core Design Contradiction:
Use of energy by moving objectVSEase of operation

Solution Approach 1:

The system predicts which users are likely to speak next based on conversation analysis and proactively sends them higher quality video feeds in advance. This ensures seamless transitions when users become active speakers, improving user experience while maintaining resource efficiency through selective prediction-based quality adjustment.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts video quality based on real-time conversation context and predicted speaker transitions. By continuously analyzing conversation patterns and adapting quality levels proactively, the system optimizes both resource usage and user experience, ensuring high quality feeds are available exactly when needed without unnecessary delays.

Inventive Principle:
Principle #15Dynamics

3Ease of operation

If the system switches video feed quality immediately when a non-speaking user becomes a speaking user, then user experience is improved, but network and system resources are wasted during the transition period

Engineering Contradiction:
Improveuser experienceVSAvoidnetwork resource waste
Core Design Contradiction:
Ease of operationVSLoss of energy

Solution Approach 1:

The system uses predictive analysis to determine which users are likely to speak next based on conversation patterns, and proactively sends higher quality video feeds only to those predicted speakers. This approach improves user experience by eliminating switching delays while avoiding resource waste by sending high quality feeds only when and where needed, based on intelligent prediction rather than blanket switching.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS8976223B1Speaker switching in multiway conversation
Publication Date: 2015.03.10 GOOGLE LLC
  • US8976223B1 patent drawing
  • US8976223B1 patent drawing
  • US8976223B1 patent drawing

AI summary

A method, computer program product, and computer system for sending, by a first computing device, a video feed with a pre-determined quality level to a second computing device. The first computing device determines that a volume level associated with the video feed reaches a threshold. The first computing device sends the video feed with a higher quality level to the second computing device based upon, at least in part, determining that the volume level associated with the video feed reaches the threshold.