Virtual Microphone Beamforming for Underdetermined Source Separation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing independent component analysis (ICA) methods are limited in separating sound source signals of a number greater than the number of microphones, leading to increased costs for sound source separation.

Innovation Solution

A sound source separation system using a configuration of n microphones and a virtual microphone signal generator to produce m virtual unidirectional microphones with different directivities, where m > n, allowing the ICA processor to separate L sound sources (L ≤ m) effectively, where L > n.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the number of microphones is increased to separate more sound sources, then the separation capability is improved, but the cost and device complexity increase

Engineering Contradiction:
Improvesound source separation capabilityVSAvoidnumber of microphones
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent creates virtual microphone signals that copy the function of additional physical microphones. By generating m virtual unidirectional microphone signals from n physical omnidirectional microphones (where m > n), the system achieves the separation capability of having more microphones without actually installing them, thus resolving the contradiction between separation capability and device complexity

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent transitions from the spatial dimension (adding more physical microphones) to the signal processing dimension (generating virtual microphone signals through beamforming and ICA). This dimensional shift allows the system to achieve higher separation capability without increasing the physical microphone count, thereby resolving the contradiction

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If the number of microphones is increased to separate more sound sources, then the separation capability is improved, but the cost increases

Engineering Contradiction:
Improvesound source separation capabilityVSAvoidcost
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent uses signal processing to create virtual copies of microphone functions. By generating m virtual unidirectional microphone signals from n physical omnidirectional microphones, the system achieves the separation capability of having more expensive physical microphones without actually purchasing them, thus resolving the contradiction between separation capability and cost

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical solution of adding more physical microphones with an electronic signal processing solution. The beamforming and ICA processing substitute for additional hardware, reducing the cost while maintaining or improving sound source separation capability

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4672239A1Sound source separation system
Publication Date: 2025.12.31 ALPS ALPINE CO LTD
  • EP4672239A1 patent drawingFigure 1
  • EP4672239A1 patent drawingFigure 2A~2B
  • EP4672239A1 patent drawingFigure 3A~3B

AI summary

A sound source separation system is configured to perform blind source separation of individual sound source signals by an independent component analysis (ICA) method from a plurality of mixed signals in which two or more sound source signals are mixed. The sound source separation system includes an n number of microphones, where n≥2; a virtual microphone signal generator configured to generate, from output signals of the n number of the microphones, an m number of the virtual microphone signals that are output signals of the m number of virtual unidirectional microphones having directivities in different directions, where m>n; and an ICA processor configured to separate an L number of the sound source signals that are signals of the L number of different sound sources, by the ICA method, from the m number of the virtual microphone signals that are the plurality of the mixed signals, where L>n and L≤m.