Audio Signal Direction Detection for Voice Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing voice recognition systems face challenges in accurately determining a target sound source among multiple external sound sources, particularly in noise environments, leading to deteriorated performance when controlling devices like TVs and Bluetooth audio devices.

Innovation Solution

An electronic device equipped with multiple microphones and a processor that determines the direction of sound sources using angle and phase information, separates audio signals, and transmits only the signal from the target sound source to a voice recognition server, enabling effective voice recognition by identifying the target sound source based on the duration of sound source directions and wake-up words.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If microphone-array technique is used to strengthen sound source and eliminate noise, then voice recognition performance is improved, but device complexity increases

Engineering Contradiction:
Improvevoice recognition performanceVSAvoiddevice complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the audio signal processing into distinct functional modules: a sound source direction determination unit that identifies directions of multiple sound sources, and a voice signal extraction unit that separates target voice from noise based on directional information. This modular segmentation improves voice recognition in noise environments while maintaining manageable system complexity through functional decomposition.

Inventive Principle:
Principle #1Segmentation

2Adaptability or versatility

If additional electronic device with microphone-array is used for remote control, then control capability is extended, but voice recognition performance deteriorates due to audio output from controlled device

Engineering Contradiction:
Improveremote control capabilityVSAvoidvoice recognition performance
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent segments the audio field into multiple directional components by determining sound source directions and separating audio signals according to these directions. This allows the system to isolate and eliminate audio output from controlled devices (TV, Bluetooth audio) while preserving the user's voice commands, thereby maintaining remote control capability without voice recognition deterioration.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary processing stage between audio signal reception and voice recognition. The sound source direction determination unit and voice signal extraction unit act as intermediaries that analyze and separate audio signals before transmission to the voice recognition engine, enabling the system to handle remote control scenarios with audio interference.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Ease of operation

If voice recognition is performed in noise environment with low SNR, then system usability is maintained, but recognition accuracy deteriorates

Engineering Contradiction:
Improvesystem usabilityVSAvoidrecognition accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent segments the mixed audio signal into multiple directional audio signals based on determined sound source directions. By separating the target voice signal from background noise through directional segmentation, the system maintains usability in noise environments while significantly improving recognition accuracy through enhanced signal isolation.

Inventive Principle:
Principle #1Segmentation

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 approach allows for stable and accurate voice recognition in various noise environments, reducing computational complexity and improving recognition accuracy by isolating the target sound source and reducing noise interference.

Implementation Method 1

determining the direction in which each of the multiple sound sources is located with reference to the electronic device, on the basis of the multiple audio signals received through the multiple microphones

Methodology Applied
Scientific EffectPhase information:

Data Source

PatentEP3480820B1Electronic device and method for processing audio signals
Publication Date: 2021.05.19 SAMSUNG ELECTRONICS CO LTD
  • EP3480820B1 patent drawingFigure 1
  • EP3480820B1 patent drawingFigure 2
  • EP3480820B1 patent drawingFigure 3

AI summary

An electronic device is disclosed. The electronic device comprises: multiple microphones for receiving audio signals generated by multiple sound sources; a communication unit for communicating with a voice recognition server; and a processor for determining the direction in which each of the multiple sound sources is located with reference to the electronic device, on the basis of the multiple audio signals received through the multiple microphones, determining at least one target sound source among the multiple sound sources on the basis of the duration of the determined direction of each of the sound sources, and controlling the communication unit such that the communication unit transmits, to the voice recognition server, an audio signal of a target sound source from which a predetermined voice is generated among the at least one target sound source.