Microphone Transfer Characteristic Adaptation for Sound Separation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The separation accuracy of target and non-target sounds is reduced when the surrounding environment changes, causing a reversal in the frequency spectrums and signal levels, which hampers successful sound analysis.

Innovation Solution

A sound processing device is coupled to multiple input devices, generating frequency spectrums, calculating transfer characteristics, and suppressing intensity based on these characteristics to adapt to environmental changes and accurately separate target sounds from non-target sounds.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If binary masking is used for sound separation, then the target sound can be separated from non-target sound under stable environmental conditions, but the separation accuracy deteriorates when the surrounding environment changes

Engineering Contradiction:
Improvesound separation accuracyVSAvoidenvironmental adaptability
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent applies parameter changes by dynamically adjusting the transfer characteristic parameters based on environmental conditions. The system estimates transfer characteristics between microphones and adapts the sound separation algorithm parameters according to the estimated environmental parameters, enabling accurate separation across varying environments without requiring manual recalibration

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent implements feedback by using the estimated transfer characteristics to continuously adjust the sound separation process. The system estimates the transfer characteristic between microphones, uses this information to adapt the binary masking parameters, and applies the adapted parameters to improve separation accuracy in real-time under changing environmental conditions

Inventive Principle:
Principle #23Feedback

2Measurement precision

If the frequency spectrum comparison is performed without environmental compensation, then the processing complexity is low, but the separation accuracy is reduced due to environmental changes

Engineering Contradiction:
Improvesound separation accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by estimating the transfer characteristic between microphones before performing the sound separation operation. This preliminary estimation of environmental parameters allows the system to pre-adjust the separation algorithm, ensuring accurate separation without adding significant complexity to the main processing flow

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system changes the parameters of the sound separation algorithm based on the estimated transfer characteristic. By dynamically adjusting parameters such as the masking threshold and frequency weighting based on environmental conditions, the system achieves high separation accuracy while maintaining computational efficiency through parameter adaptation rather than complex reprocessing

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3291227B1Sound processing device, method of sound processing, sound processing program and storage medium
Publication Date: 2019.09.18 FUJITSU LTD
  • EP3291227B1 patent drawingFigure 1
  • EP3291227B1 patent drawingFigure 2
  • EP3291227B1 patent drawingFigure 3

AI summary

A sound processing device includes a processor configured to generate a first frequency spectrum of a first sound signal corresponding to a first sound received at a first input device and a second frequency spectrum of a second sound signal corresponding to the first sound received at a second input device, calculate a transfer characteristic based on a first difference between an intensity of the first frequency spectrum and an intensity of the second frequency spectrum, generate a third frequency spectrum of a third sound signal transmitted from the first input device and a fourth frequency spectrum of a fourth sound signal transmitted from the second input device, specify a suppression level of an intensity of the fourth frequency spectrum based on a second difference between an intensity of the third frequency spectrum, an intensity of the fourth frequency spectrum.