Voice Latency Estimation for Conference Collision Avoidance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Audio or video conferencing systems experience disrupted communication due to inevitable delays in audio/video transmission, leading to uncertainty for users about when others have perceived speech offsets, increasing the likelihood of collisions during conversations.

Innovation Solution

A client device system that includes an offset detecting unit, a configuring unit, and an estimator to determine voice latency and estimate when speech offsets are perceived by other parties, with output units providing perceivable signals to inform users when others have detected these offsets, helping to mitigate delays and collisions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If audio/video transmission is performed in conference systems, then communication between parties at different places is enabled, but transmission delays cause disrupted communication and increase the likelihood of collisions

Engineering Contradiction:
Improvecommunication capabilityVSAvoidcommunication stability
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The system provides feedback to the user about the transmission delay state and collision risk. The processing unit calculates delay parameters and the output unit presents this information to the user, enabling the user to adjust their speech timing accordingly and avoid collisions with far-end parties.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs preliminary calculation of transmission delay parameters before the user speaks. By providing advance information about the expected delay and collision risk, the user can take preliminary action to time their speech appropriately, preventing collisions before they occur.

Inventive Principle:
Principle #10Preliminary action

2Device complexity

If transmission delays are not compensated, then system complexity remains low, but users experience uncertainty about when others perceive speech offsets

Engineering Contradiction:
Improvesystem simplicityVSAvoidtiming information
Core Design Contradiction:
Device complexityVSLoss of information

Solution Approach 1:

The system calculates transmission delay parameters and provides feedback to the user about when their speech will be perceived by far-end parties. This feedback loop restores the timing information that is lost due to transmission delays, allowing users to speak at appropriate times without complex end-to-end delay compensation.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

Each client device independently calculates its own transmission delay parameters and provides timing information to its local user. This self-service approach avoids the need for complex centralized delay compensation systems while still providing the necessary timing information to prevent collisions.

Inventive Principle:
Principle #25Self-service

3Reliability

If end-to-end delay compensation is implemented, then communication timing can be synchronized, but system complexity and computational requirements increase significantly

Engineering Contradiction:
Improvetiming synchronizationVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts the delay compensation function from the complex end-to-end system and implements it locally at each client device. Each device independently calculates its own transmission delay parameters and provides timing information to its user, eliminating the need for complex centralized delay compensation while achieving the same timing synchronization effect.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Each client device performs self-service delay compensation by independently calculating its transmission delay parameters and providing timing information to its local user. This distributed approach avoids the computational complexity of end-to-end delay compensation while achieving effective timing synchronization.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2901669B1Near-end indication that the end of speech is received by the far end in an audio or video conference
Publication Date: 2018.07.18 DOLBY INTERNATIONAL AB
  • EP2901669B1 patent drawingFigure 1~2B
  • EP2901669B1 patent drawingFigure 2C~2D
  • EP2901669B1 patent drawingFigure 3

AI summary

Embodiments of client device and method for audio or video conferencing are described. An embodiment includes an offset detecting unit, a configuring unit, an estimator and an output unit. The offset detecting unit detects an offset of speech input to the client device. The configuring unit determines a voice latency from the client device to every far end. The estimator estimates a time when a user at the far end perceives the offset based on the voice latency. The output unit outputs a perceivable signal indicating that a user at the far end perceives the offset based on the time estimated for the far end. The perceivable signal is helpful to avoid collision between parties.