3D Audio Conference Speaker Positioning via Voice Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio conferencing systems lack efficient methods for automatically and optimally positioning speakers in a 3D virtual space, leading to poor listening comfort and intelligibility, especially as the number of participants increases, and often require manual or random positioning.

Innovation Solution

A method that estimates specific characteristics from digital signals, such as voice features or terminal information, to determine optimal virtual positions for speakers, eliminating the need for manual interfaces and ensuring optimal positioning regardless of the number of participants.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual positioning of speakers is implemented, then speakers can be positioned in 3D virtual space, but the complexity of operation increases and optimal positioning becomes difficult beyond a few participants

Engineering Contradiction:
Improvespeaker positioning operationVSAvoidpositioning interface complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The system automatically positions speakers by analyzing voice characteristics and computing optimal positions without requiring user interaction. The computer resources estimate voice characteristics from digital signals and determine positioning instructions autonomously, eliminating the need for manual positioning interfaces and operations.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The manual mechanical process of positioning speakers through user interface interactions is replaced by an automated computational system that analyzes voice characteristics and computes optimal positions algorithmically, substituting human operation with automated signal processing and position calculation.

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

2Ease of operation

If random positioning of speakers is used, then positioning is simple to implement, but listening comfort and intelligibility deteriorate

Engineering Contradiction:
Improvepositioning implementation simplicityVSAvoidlistening comfort and intelligibility
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The system uses feedback from voice characteristic analysis to determine optimal speaker positions. By estimating characteristics from digital signals and using this information to compute positioning instructions, the system adapts positions based on actual voice properties rather than using random placement, thereby improving listening comfort and intelligibility.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system changes the positioning parameters from random values to optimized values based on voice characteristics. By analyzing voice properties and computing positions that maximize spatial separation and listening quality, the system transforms arbitrary positioning into scientifically optimized positioning while maintaining automated operation.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If the number of speakers increases, then more participants can join the conference, but optimal positioning becomes increasingly difficult to guarantee

Engineering Contradiction:
Improveconference capacityVSAvoidpositioning optimization quality
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The automated system handles positioning for any number of speakers without requiring additional user effort or interface complexity. Computer resources automatically estimate voice characteristics and compute optimal positions for all participants, scaling the solution to accommodate increasing conference capacity while maintaining positioning quality.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The positioning system dynamically adapts to changing conference conditions as speakers join or leave. By continuously analyzing voice characteristics and recalculating optimal positions, the system maintains high-quality spatial audio configuration regardless of the number of participants, making the positioning quality resilient to changes in conference size.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP2009892B1Positioning of speakers in a 3-D audio conference
Publication Date: 2019.03.06 ORANGE SA
  • EP2009892B1 patent drawingFigure 1
  • EP2009892B1 patent drawingFigure 2
  • EP2009892B1 patent drawingFigure 3

AI summary

The method involves estimating a proper characteristic of a speaker from a digital signal issued from a terminal of the speaker. A set point for positioning the speaker in a virtual space of listener is determined from the estimated characteristic. A determined set point is delivered by an output. The speaker is virtually spatialized in the virtual space of the listener by using the determined set point. Independent claims are also included for the following: (1) a device for establishing an audio conference between the instructors (2) a computer program stored in a memory of the device.