Hearing Aid Noise Suppression Using Adaptive Filter Difference Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing binaural hearing aid systems suffer from sound artifacts during reverberation suppression, impairing speech intelligibility and listening comfort for users.
Innovation Solution
A method utilizing a time-varying adaptive filter to derive precise noise estimates from difference signals between two spatially separated input transducers, which are then used in noise reduction algorithms like spectral subtraction to minimize processing artifacts and effectively suppress both correlated and uncorrelated noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If reverberation suppression is applied in binaural hearing aid systems, then speech intelligibility should be improved, but sound artifacts are generated that impair speech intelligibility and listening comfort
Solution Approach 1:
The patent segments the noise estimation process into two distinct components: correlated noise estimation using adaptive filtering and uncorrelated noise estimation using statistical methods. This segmentation allows each estimator to specialize in specific noise types, reducing artifacts while maintaining speech intelligibility. The correlated noise path processes signals through adaptive filters that track environmental changes, while the uncorrelated noise path handles random background sounds, together providing comprehensive noise suppression without generating harmful artifacts.
Solution Approach 2:
The patent introduces an intermediary mechanism that combines outputs from multiple noise estimators (correlated and uncorrelated) before applying noise suppression. This intermediary combination stage allows the system to leverage the strengths of different estimation methods while compensating for their individual weaknesses, thereby reducing artifacts. The intermediary also implements artifact detection and suppression logic that monitors the processed signal and removes harmful artifacts while preserving speech content.
2Reliability
If noise estimation is performed continuously, then noise suppression effectiveness is improved, but processing complexity and energy consumption increase
Solution Approach 1:
The patent implements periodic noise estimation cycles rather than continuous processing. The system performs noise estimation at specific intervals determined by environmental stability and speech activity detection. During periods of stable acoustic environment, noise estimation is performed less frequently, reducing processing complexity. When changes are detected, the estimation frequency increases to maintain effectiveness. This periodic approach balances reliability with reduced computational burden and energy consumption.
Solution Approach 2:
The patent employs dynamic adjustment of noise estimation parameters and processing intensity based on real-time conditions. The system adapts the complexity of noise estimation algorithms according to environmental factors, speech activity levels, and current noise conditions. In simple acoustic environments, simplified estimation methods are used, while complex environments trigger more sophisticated processing. This dynamic adaptation maintains noise suppression effectiveness while optimizing processing complexity and energy usage according to actual needs.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of noise suppression in a hearing aid system by providing an improved noise estimate derived from the difference of a first digital audio signal provided by a first input transducer and an adaptively filtered second digital audio signal provided by a second input transducer. The invention further provides a hearing aid (100, 200) and a hearing aid system (300 and 400) adapted for improving noise suppression in accordance with this method.