Hearing Aid Source Separation Parameter Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hearing aid devices face challenges in distinguishing between useful and interfering signals, leading to unintended reduction of useful signals during suppression of interference, and require high computing power for source separation, which affects battery life and efficiency.
Innovation Solution
A method and system that generates parameters from source-specific received signals to emphasize or suppress signals, rather than processing them directly, allowing for reduced computing power by performing source separation only intermittently and using an external processor unit for computation-intensive tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If source separation algorithms are used to separate acoustic signals from different sound sources, then the ability to distinguish useful signals from interfering signals is improved, but the computing power required increases significantly
Solution Approach 1:
The patent performs source separation and signal classification in advance to generate parameter settings, rather than continuously processing signals in real-time. The source-specific received signals are analyzed once to determine parameters that are then reused for multiple operating states, reducing the overall computing power requirement while maintaining separation accuracy.
Solution Approach 2:
The patent extracts only the essential parameter settings from the source separation process, rather than processing all source-specific received signals continuously. By taking out only the necessary parameter information and using it to control the signal processing unit, the system achieves effective signal separation with reduced computational load.
2Reliability
If source-specific received signals are processed directly to suppress interference, then the suppression effectiveness is improved, but the computing effort increases and affects battery life
Solution Approach 1:
The patent performs the computationally intensive source separation and signal analysis in advance to generate parameter settings, which are then stored and reused. This preliminary action reduces the need for continuous high-power processing during normal operation, thereby reducing battery consumption while maintaining effective interference suppression.
Solution Approach 2:
The patent changes the operational parameters of the hearing aid device based on pre-analyzed source-specific received signals. By determining parameter settings once and applying them across multiple operating states, the system achieves effective interference suppression without the continuous energy expenditure required for real-time signal processing.
3Device complexity
If traditional noise suppression algorithms are used, then the processing complexity is reduced, but the useful signal is also reduced along with the interfering signal
Solution Approach 1:
The patent segments the mixed acoustic signal into source-specific received signals using source separation algorithms. By dividing the signal into components from different sound sources, the system can selectively process only the interfering signal components while preserving the useful signal components, thereby reducing useful signal loss while maintaining manageable processing complexity.
Data Source
Figure 1~2
Figure 3
AI summary
In a hearing aid system with at least one hearing aid (10), interference signals should be suppressed and desired signals emphasized with reasonable computational effort. For this purpose, source-specific received signals (Q1, Q2, ..., Qm; Q1', Q2', ..., Qm') are first generated and analyzed using a source separation device (3; 13), in particular a BSS unit. Parameter settings for the operation of the hearing aid (10) are then derived from the analysis results, which effect the suppression of interference signals or the emphasis of desired signals. An output signal generated by the hearing aid (10) does not arise directly from the source-specific received signals, so that time-lapse operation of the BSS unit is sufficient and real-time signal processing in the BSS unit is not necessarily required.