Direction-Dependent Noise Suppression for Hearing Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hearing devices struggle with inefficient noise suppression methods that either impair spatial perception or introduce artifacts, particularly in dynamic noise environments like 'cocktail party' situations.
Innovation Solution
A method for direction-dependent noise suppression using at least two input transducers to generate interference and target signals, with frequency-band weighting based on acoustic characteristics, generating a target signal with a homogeneous directional characteristic and attenuating interference in specific directions, ensuring a natural sound image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If directional microphone techniques are used to amplify target signal and attenuate noise, then signal-to-noise ratio is improved, but spatial perception of the wearer's surroundings is impaired
Solution Approach 1:
The patent segments the noise suppression process into multiple frequency bands, applying different weighting factors to each band based on acoustic characteristics. This allows selective noise reduction in specific frequency ranges while preserving spatial information in others, resolving the contradiction between SNR improvement and spatial perception maintenance.
Solution Approach 2:
The patent applies direction-dependent weighting factors that vary by frequency band and spatial direction. Instead of uniform noise suppression, the system tailors the suppression characteristics to specific frequency bands and directions, preserving spatial perception in certain bands while enhancing SNR in others where noise is more problematic.
2Reliability
If spectral subtraction is used to filter interfering signals, then noise reduction is achieved, but artifacts are introduced that may degrade the speech signal
Solution Approach 1:
The patent changes the approach from direct spectral subtraction to a weighting factor-based method. By adjusting frequency-band-specific weighting factors based on acoustic characteristics, the system achieves noise reduction without the aggressive signal manipulation that causes artifacts, thereby preserving speech signal quality.
3Reliability
If aggressive noise suppression is applied to improve signal clarity, then noise reduction is enhanced, but natural sound image is degraded
Solution Approach 1:
The patent dynamically adjusts weighting factors based on real-time acoustic characteristics of target and interference signals. This dynamic adaptation allows the system to achieve effective noise reduction while maintaining the natural sound image, as the weighting is optimized for each acoustic situation rather than applying fixed aggressive suppression.
Data Source
Figure 1A~1B
Figure 2
Figure 3~4
AI summary
The invention relates to a method for direction-dependent noise suppression for a hearing system (2), which comprises a hearing device (1), wherein, by means of at least one first input transducer (21) of the hearing system (2) and a second input transducer (22) of the hearing system (2), an interference signal (36) and a target signal (40) are generated from ambient sound, wherein the interference signal (36) and/or the target signal (40) are referenced to a useful signal source arranged in a target direction (38), wherein the target signal (40) is generated with a target directional characteristic which is homogeneous or substantially homogeneous over a half-space (66) opposite the target direction (38), wherein, for at least a first plurality of frequency bands, an acoustic characteristic (42, 46) of the target signal (40) is compared with a corresponding acoustic characteristic (42, 44) of the interference signal (36).and based on the aforementioned comparison, a preliminary weighting factor (51) is determined, the range of values (80) of which has at least three values (80a, 80b, 80), wherein, for each frequency band, a weighting factor (54) is formed based on the preliminary weighting factor (51), and wherein an input signal (56) of the hearing system (2) to be processed is weighted frequency band by frequency band based on the respective weighting factor (54), and an output signal (58) is generated based on the input signal (56) to be processed in this way.