Sound Source Separation via 2D Fourier Transform
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sound source separation technologies fail to effectively utilize characteristic sound collection results when sound sources are positioned at equal distances from multiple microphones, leading to incomplete separation of overlapping sound spectra.
Innovation Solution
A sound source separation system that acquires and processes sound collection data using a controller to distinguish between sound sources by performing Fourier transforms in two dimensions, leveraging the unique phase characteristics of sounds collected at equal distances from microphones, allowing for clear separation of overlapping spectra.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sound sources are positioned at equal distances from multiple microphones, then the sound pressure is equal across all microphones, but the existing technology fails to effectively utilize this characteristic for separation
Solution Approach 1:
The patent transforms the sound collection data from one-dimensional time-domain signals into two-dimensional frequency spectra by performing Fourier transforms in both the time direction and the circumferential direction of the microphone array. This dimensional transformation enables the system to exploit the characteristic that sounds from equal-distance positions concentrate at zero frequency in the circumferential direction, while sounds from other positions distribute across multiple frequencies, thereby achieving effective separation.
2Reliability
If traditional sound source separation methods are used, then the process is simple, but they cannot separate sound sources when they are present at positions with equal distances from microphones
Solution Approach 1:
The patent changes the processing parameters from simple time-domain signal processing to two-dimensional frequency spectrum analysis. By performing Fourier transforms in both time and circumferential directions, the system transforms the sound data into a parameter space where equal-distance sound sources exhibit distinct spectral characteristics (concentration at zero circumferential frequency), enabling reliable separation that was not achievable with traditional methods.
3Measurement precision
If sound sources are separated by collecting sounds from multiple microphones, then the sound pressure distribution can be obtained, but the characteristic results from equal-distance positions are not fully utilized
Solution Approach 1:
The patent introduces a feedback mechanism where the two-dimensional frequency spectrum analysis provides information about the circumferential direction characteristics back to the sound source position estimation process. The system uses the concentrated energy at zero circumferential frequency (which indicates equal-distance positioning) as feedback to identify and separate sound sources, thereby fully utilizing the phase characteristic information that was previously underutilized.
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
The system effectively separates sound spectra by exploiting the phase differences, enabling accurate identification and separation of sound sources even when they are present together, improving noise attenuation and source localization accuracy.
Implementation Method 1
collecting sounds output from a first sound source present at a first position at which effective distances from a plurality of microphones are equal and a second sound source present at a position other than the first position by the plurality of microphones
Implementation Method 2
when Fourier transform is performed on such pieces of sound collection data in the two dimensions of the circumferential direction and the time direction
Data Source
AI summary
A sound source separation system includes: a controller that: acquires pieces of sound collection data with microphones that collect sounds output from first and second sound sources. The first sound source is at a first position at which effective distances from the microphones are equal and the second sound source is at a different position. The controller further acquires, based on the sound collection data, frequency spectra in two dimensions of a circumferential direction of a circle and a time direction. The first position is a center of the circle and each of the effective distances is a radius of the circle. The controller further separates, from the frequency spectra, a first sound source spectrum and a second sound source spectrum.