Directional Sound Source Filtering Using Microphone Array Beamforming

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing digital devices face challenges in selectively amplifying sound sources within an image area while effectively eliminating sound from outside areas, leading to noise and interference due to environmental factors and the need for image processing for face recognition.

Innovation Solution

A directional sound source filtering apparatus using a microphone array, which includes an image detector, a sound collector, and a controller that calculates time delay values to separate sound sources within and outside the image area, employing beamforming and frequency conversion to isolate and amplify only infield sound sources, and uses a least mean square filter to eliminate non-directional noise.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If sound source signals are captured using microphones in digital devices, then sound sources can be detected, but noise and interference sound are introduced due to environmental factors

Engineering Contradiction:

Improvesound source detection accuracyVSAvoidnoise and interference sound

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent divides the spatial domain into image area and non-image area based on camera viewing angle information. The microphone array is configured to detect sound sources in different directions, and the controller separates sound source signals into those corresponding to the image area and those corresponding to the non-image area, thereby eliminating noise and interference sound from the audio signal

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces camera viewing angle information as an intermediary to correlate audio and video spatial information. This intermediary parameter enables the controller to identify which sound source signals correspond to the image area and which correspond to the non-image area, allowing for effective separation of relevant audio from irrelevant noise

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If face recognition image processing is used to discern speaker location, then selective amplification of sound sources can be achieved, but the performance is affected by image processing complexity

Engineering Contradiction:

Improveselective amplification performanceVSAvoidimage processing performance

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts the spatial location information directly from camera viewing angle data without performing complex face recognition processing. The controller uses the viewing angle information to determine which sound source signals correspond to the image area, thereby achieving selective amplification without the computational complexity of face recognition algorithms

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces the image processing-based face recognition system with a microphone array-based acoustic field analysis system. Instead of using visual recognition to locate speakers and then filter audio, the system uses spatial filtering of acoustic signals directly, substituting a complex visual processing mechanism with a more efficient acoustic signal processing approach

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

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 enhances the capability to selectively amplify sound sources within the image area, reducing noise and interference, and improving the signal-to-interference ratio by effectively filtering out non-directional noise, thus providing clearer audio transmission.

Implementation Method 1

the controller precalculate time delay values of sound sources within the images to extract sound sources within the image from the sound sources detected by the sound collector

Methodology Applied

Scientific EffectTime delay:

Implementation Method 2

perform beamforming through the calculated time delay values

Methodology Applied

Scientific EffectBeamforming:

Data Source

PatentUS9226070B2Directional sound source filtering apparatus using microphone array and control method thereof

Publication Date: 2015.12.29 SAMSUNG ELECTRONICS CO LTD

AI summary

A directional sound source filtering apparatus using a microphone array and a control method thereof are provided. The directional sound source filtering apparatus using a microphone array includes an image detector to detect images in a destination area, a sound collector located by the microphone array in which microphones are arranged to detect sound sources together with the images detected by the image detector. The apparatus includes a controller to precalculate time delay values of sound sources within the images in order to extract sound sources within the image from the sound sources detected by the sound collector, and perform beamforming through the calculated time delay values. Sound source signals only within images may be selectively amplified using beamformers.