Spatial Filterbank for Hearing System Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional hearing systems, such as hearing aids, struggle to effectively differentiate between sound sources in a noisy environment, leading to poor signal-to-noise ratios and difficulties in localizing sound sources, especially in dynamic acoustic scenarios.
Innovation Solution
A hearing system equipped with a spatial filterbank that divides the acoustic environment into subspaces, allowing for improved voice signal detection, noise reduction, and sound source localization by processing sound signals from specific directional subspaces, enabling adaptive adjustment of subspace parameters for enhanced auditory experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a beamformer is used to suppress sound from other directions, then the signal-to-noise ratio is increased, but the spatial resolution and ability to differentiate between multiple sound sources is limited
Solution Approach 1:
The spatial filterbank divides the total acoustic space into multiple frequency-dependent subspaces, each processed by separate filterbank channels. This segmentation allows simultaneous beamforming in multiple directions across different frequency bands, improving both signal-to-noise ratio and spatial resolution by treating different spatial-frequency components independently
Solution Approach 2:
The patent extends conventional single-channel beamforming into a multi-dimensional framework by introducing frequency-band-specific spatial filtering. Each filterbank channel processes a specific frequency range with its own spatial characteristics, adding a frequency dimension to the spatial filtering operation and enabling more precise sound source differentiation
2Area of stationary object
If omnidirectional microphones are used to receive sound from all directions, then the coverage area is maximized, but the ability to localize sound sources and suppress noise is lost
Solution Approach 1:
The spatial filterbank segments the omnidirectional sound field into multiple directional subspaces across different frequency bands. Each filterbank channel captures sound from specific spatial directions within its frequency range, maintaining overall coverage while enabling precise sound source localization through the combined output of multiple segmented channels
Solution Approach 2:
Different filterbank channels are assigned different spatial characteristics tailored to their frequency ranges. High-frequency channels may focus on narrower directional beams while low-frequency channels maintain broader coverage, optimizing both localization precision and overall sound reception coverage for each frequency band
3Reliability
If a single beamformer is used to focus on one direction, then the signal-to-noise ratio for that direction is improved, but the system cannot quickly switch between multiple sound sources
Solution Approach 1:
The spatial filterbank creates multiple parallel beamforming channels, each optimized for different spatial directions and frequency bands. The system can rapidly switch between sound sources by selectively activating different filterbank channels without requiring time-consuming beam reformation, significantly improving switching speed while maintaining high signal-to-noise ratio through the segmented channel architecture
Solution Approach 2:
The system dynamically selects and combines outputs from multiple filterbank channels based on real-time sound source positions and priorities. This dynamic channel selection enables rapid adaptation to changing acoustic environments and quick switching between sound sources while maintaining optimal signal-to-noise ratio through the pre-configured segmented channels
Data Source
Figure 1
Figure 2A~2B
Figure 2C~2E
AI summary
The present invention regards a hearing system (10) configured to be worn by a user (62) comprising an environment sound input unit (12, 14), an output transducer (18), and electric circuitry (16). The environment sound input unit (12, 14) is configured to receive sound (20) from the environment of the environment sound input unit (12, 14) and to generate sound signals (22, 24) representing sound (20) of the environment. The output transducer (18) is configured to stimulate hearing of a user (62). The electric circuitry (16) comprises a spatial filterbank (34). The spatial filterbank (34) is configured to use the sound signals (22, 24) to generate spatial sound signals (56) dividing a total space (60) of the environment sound (20) in subspaces (58). Each spatial sound signal (56) represents sound (20) coming from a subspace (58). The subspaces may (in particular modes of operation) be either fixed, or dynamically determined, or a mixture thereof.