Real-Time Blind Sound Source Separation Using Auxiliary Function Approximation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional blind sound source separation techniques using independent component analysis face challenges in achieving real-time processing while adapting to environmental changes, such as movement of sound sources, due to high computational requirements and the need for extensive parameter adjustments.
Innovation Solution
A signal processing apparatus that employs an auxiliary function method to estimate an auxiliary variable and update the demixing matrix in real-time, reducing the need for extensive observation signal referencing and minimizing computational load by using an approximating auxiliary function to separate signals efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional independent component analysis is used for blind sound source separation, then signal separation accuracy can be achieved, but the computational load increases and real-time processing becomes difficult
Solution Approach 1:
The patent pre-calculates and stores the auxiliary variable based on historical observation signals before they are needed for separation. This preliminary computation allows the separation process to proceed quickly using pre-computed values, resolving the contradiction between accurate separation and real-time processing speed
Solution Approach 2:
The patent creates an auxiliary variable that copies and transforms the essential statistical properties of the observation signals. This auxiliary variable serves as a simplified representation that can be computed once and reused multiple times, reducing the computational burden during actual separation operations while maintaining accuracy
2Measurement precision
If conventional methods reference extensive observation signals for accurate separation, then separation accuracy improves, but the computational burden increases
Solution Approach 1:
The patent extracts only the essential statistical features from the observation signals to create the auxiliary variable, rather than processing the entire signal dataset. This extraction approach captures the necessary information for accurate separation while discarding redundant data, thereby reducing computational burden
Solution Approach 2:
The patent uses a partial representation (the auxiliary variable) instead of the complete observation signal set. This partial action provides sufficient information for accurate separation without requiring exhaustive processing of all available signals, optimizing the balance between accuracy and computational efficiency
Data Source
AI summary
According to an embodiment, a signal processing apparatus includes an estimation unit and an updating unit. The estimation unit is configured to estimate an auxiliary variable of a target section including first and second sections of input signals by using an approximating auxiliary function for approximating an auxiliary function having an auxiliary variable as an argument. The auxiliary function is determined according to an objective function that outputs a function value that is smaller as a statistical independence of separated signals into which input signals in time-series are separated by a demixing matrix is higher. The estimation unit is configured to estimate a value of the auxiliary variable of the target section based on the estimated auxiliary variable. The updating unit is configured to update the demixing matrix such that a function value of the approximating auxiliary function is minimized.