Speaker Isolation via Sound Localization and Dynamic Muting

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current voice communication technologies, such as telephones and vehicles' built-in communication systems, face challenges in distinguishing individual speakers amidst background noise and ensuring clear communication, particularly during conferences and one-speaker calls, leading to distractions and poor user experience.

Innovation Solution

A voice communication apparatus equipped with multiple sound sensors and a mute control system that analyzes sound characteristics, including position, phase, and voice analysis, to dynamically mute or unmute based on the presence and location of individual speakers, ensuring only the intended speaker is heard.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple microphones are used to capture speech from individual speakers, then the ability to hear individual speakers is improved, but background noise from other speakers and directions is also captured

Engineering Contradiction:
Improvespeaker identification accuracyVSAvoidbackground noise interference
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent divides the audio capture function into multiple spatially distributed microphones, each responsible for capturing sound from specific directions. The system segments the audio field into multiple zones and assigns different microphones to monitor different zones, allowing selective amplification of desired speakers while attenuating background noise from other directions

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different processing characteristics to different spatial zones. Each microphone is optimized for its specific directional zone, and the system applies localized noise filtering and gain control based on the spatial position of sound sources. This allows the system to enhance speech from target speakers while selectively suppressing background noise from other directions

Inventive Principle:
Principle #3Local quality

2Measurement precision

If the system automatically switches microphones in response to speaker direction, then individual speaker clarity is improved, but quiet speakers are missed when louder speakers are present

Engineering Contradiction:
Improvespeaker selection accuracyVSAvoidspeaker detection reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where the system continuously monitors audio signals from all microphones and adjusts microphone selection based on real-time analysis. The system provides feedback to identify when a quiet speaker is present despite louder competitors, and automatically adjusts gain and selection to ensure the quiet speaker is captured when they speak

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent employs dynamic microphone selection and gain control that adapts to changing speaker conditions. The system continuously analyzes speaker characteristics and adjusts its behavior in real-time, allowing it to switch between automatic direction-based selection and voice-activity-based selection depending on the current acoustic environment and speaker conditions

Inventive Principle:
Principle #15Dynamics

3Object-affected harmful factors

If the user manually mutes and unmutes the phone during calls, then background noise is reduced, but user convenience and experience deteriorate

Engineering Contradiction:
Improvebackground noise controlVSAvoiduser convenience
Core Design Contradiction:
Object-affected harmful factorsVSEase of operation

Solution Approach 1:

The patent implements automatic noise control that operates without user intervention. The system autonomously identifies background noise sources, determines when muting is appropriate based on active speaker detection and noise level analysis, and executes muting/unmuting decisions automatically, freeing the user from manual control while maintaining noise suppression

Inventive Principle:
Principle #25Self-service

4Measurement precision

If sound sensors are used to detect speaker position, then individual speaker isolation is improved, but system complexity increases

Engineering Contradiction:
Improvespeaker position detection accuracyVSAvoidsensor and processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent makes the existing microphone array perform multiple functions: general audio capture, directional speaker identification, and position-based noise suppression. By processing the same audio signals through multiple algorithms simultaneously, the system achieves precise speaker isolation without adding dedicated position detection sensors, thereby avoiding increased hardware complexity

Inventive Principle:
Principle #6Universality (Multi-functionality)

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

This solution effectively reduces background noise interference and enhances communication clarity by accurately identifying and isolating individual speakers, improving both conference calls and one-speaker calls, and addressing the issues of background noise in vehicles.

Implementation Method 1

sensing the position of individual speakers can comprise at least one of: comparing individual speaker sound loudness between sound sensors; comparing individual speaker sound phase between sound sensors; and comparing of individual speaker sound arrival time between sound sensors

Methodology Applied
Scientific EffectSound wave propagation: Sound

Data Source

PatentEP2883344B1Automatic call muting method and apparatus using sound localization
Publication Date: 2018.12.12 CHINTALA SANDEEP KUMAR
  • EP2883344B1 patent drawingFigure 1
  • EP2883344B1 patent drawingFigure 2A~2D
  • EP2883344B1 patent drawingFigure 3

AI summary

A mobile telephone device (10) comprises a microphone (20) and one or more spaced audio sensors (12) to sense the position and/or voice characteristics of one or more individual speakers (16). A mute control means (14) employs sound phase, and/or sound time of arrival, and/or sound loudness to create a map of the positions of individual speakers (16). The mute control means (14) identifies individual speaker (16) voice characteristics using one, the other or both of audio signal analysis of the sound of individual speaker's voices; and use of voice CODEC analysis results for each individual speaker. A call may involve no sound muting, may involve sound muting except for one individual speaker (16), or sound muting except for any one of a plurality of accepted individual speakers. A default individual speaker position immediately before and closest the microphone is provided. Positional tolerance for individual speakers of at least 5% to 10% is employed.