Hearing Aid Feedback Detection Using Dual Algorithm Units
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hearing devices face challenges in reliably and rapidly detecting acoustic feedback, leading to annoying whistling due to the trade-off between detection speed and accuracy in existing solutions.
Innovation Solution
The implementation of a dual feedback detection unit system with different algorithms and a comparison unit to determine a threshold value, allowing for optimized feedback suppression by signaling when the threshold is exceeded, enabling adaptation to the user's characteristics and habits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a large increment is used for rapid adaptation of filter coefficients, then the speed of feedback suppression is improved, but signal artifacts are generated
Solution Approach 1:
The patent applies dynamics by making the increment adaptive rather than fixed. The increment is automatically adjusted based on the detected feedback situation - large when feedback is present to enable rapid suppression, and small or zero when feedback is absent to avoid generating signal artifacts. This dynamic adjustment resolves the contradiction between speed and artifact generation.
Solution Approach 2:
The patent changes the parameter of increment size based on the feedback detection result. When feedback is detected, the increment is increased to accelerate adaptation; when feedback is not detected, the increment is reduced to prevent artifacts. This parameter change strategy directly addresses the trade-off between adaptation speed and artifact generation.
2Measurement precision
If detection accuracy is improved using reliable methods, then feedback detection precision is improved, but detection speed decreases
Solution Approach 1:
The patent segments the feedback detection process into two distinct units: a first feedback detection unit for rapid but less accurate detection, and a second feedback detection unit for slower but more accurate detection. This segmentation allows the system to leverage the strengths of both approaches - speed from the first unit and accuracy from the second unit - without being constrained by the trade-off in a single detection method.
Solution Approach 2:
The patent merges the outputs of two different feedback detection algorithms with complementary characteristics. The first detection unit provides quick initial detection, while the second detection unit provides refined accurate detection. By combining these two detection streams, the system achieves both rapid response and high accuracy in feedback detection.
3Device complexity
If a fixed increment is used for feedback suppression, then device complexity is reduced, but adaptability to different feedback situations deteriorates
Solution Approach 1:
The patent implements self-service by enabling the feedback suppression system to automatically adjust its own increment based on real-time feedback detection. The system monitors the feedback situation and autonomously modifies the increment parameter without requiring external intervention or complex manual configuration, thus maintaining simplicity while achieving high adaptability.
Solution Approach 2:
The patent uses feedback from the detection units to control the increment parameter. The detection results feed back into the adaptation mechanism, creating a closed-loop system where the increment is continuously optimized based on the current feedback situation. This feedback mechanism provides adaptability without significantly increasing system complexity.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
The invention provides arrangements and associated methods for detecting (22) acoustic feedback in a hearing device (1). One embodiment comprises a first feedback detection unit (61) that determines a feedback probability (16), a second feedback detection unit (62) that determines a weighting factor (17), and a processing unit (63) that multiplies the feedback probability (16) by the weighting factor (17). Alternatively, instead of determining the weighting factor (17), a threshold value (20) can also be controlled. The invention offers the advantage of improving acoustic feedback detection (22) by combining two different feedback detection methods (61, 62).