Hearing Device Feedback-Canceller Data for Ear-Canal Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hearing devices, particularly hearing aids, often require physiological fitting by audiologists, which is not always accessible, leading to difficulties in correct positioning and user experience, especially for those purchasing devices without professional assistance.
Innovation Solution
A method utilizing a feedback canceller to determine adaptive filter coefficients, applying data manipulation and a neural network to assess proper seating of the hearing device in the ear canal, providing user feedback on positioning without relying on specialist reference values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If hearing devices require physiological fitting by audiologists, then positioning accuracy is improved, but device accessibility and ease of operation deteriorate
Solution Approach 1:
The hearing device automatically detects seating position using feedback path analysis and neural network classification, enabling users to position devices correctly without audiologist assistance. The system self-evaluates feedback data and provides real-time positioning guidance independently.
Solution Approach 2:
The system uses feedback path data from the hearing device's own acoustic feedback to determine seating position. By analyzing the feedback characteristics and comparing them against trained neural network models, the device receives feedback about its positioning accuracy and can guide users to improve it.
2Measurement precision
If complex data processing is applied to determine seating position, then measurement precision is improved, but energy consumption and processing time increase
Solution Approach 1:
The system extracts only the essential feedback path features needed for seating position detection, rather than processing complete audio signals. By focusing on specific feedback characteristics and using dimensionality reduction techniques, energy-intensive processing is minimized while maintaining detection accuracy.
Solution Approach 2:
The neural network is trained offline with extensive data, transforming complex processing requirements into a compact model that runs efficiently on the hearing device. The online processing uses pre-computed parameters and simplified calculations, reducing real-time energy consumption while preserving measurement precision.
3Measurement precision
If professional fitting assistance is required, then positioning accuracy is improved, but device complexity increases
Solution Approach 1:
The hearing device autonomously performs seating position detection and guidance without requiring external professional equipment or systems. The integrated feedback analysis and neural network classification enable the device to self-evaluate and self-guide positioning, reducing overall system complexity.
4Measurement precision
If extensive data processing is performed on feedback canceller data, then measurement precision is improved, but processing speed deteriorates
Solution Approach 1:
The system extracts only the critical features from feedback canceller data that are relevant for seating position detection. By filtering and selecting essential parameters rather than processing complete datasets, the system achieves accurate measurements with faster processing speeds.
Solution Approach 2:
The neural network is pre-trained offline with extensive feedback data, performing complex computational work before deployment. During actual operation, the pre-trained model rapidly classifies seating positions using simple inference, achieving both high precision and fast processing speed.
Data Source
Figure 1
AI summary
According to the invention a method for operation of a hearing device (2) is described. The hearing device (2) comprises a microphone (8) for the reception of ambient sound, a signal processor (10) for processing a microphone signal (SM) based on the received ambient sound into an output signal (SO), a speaker (12) for outputting the output signal (SO), the signal processor (10) comprising a feedback canceller (14) for determining and, if present, cancellation of feedback. The method comprises the steps of - retrieving for a predetermined sample period a set of data of the feedback canceller (14) that represents an acoustic feedback path between the speaker (12) and the microphone (8), - applying at least one step of reducing an amount of the data within the set of data, - applying a denoising step to the set of data, - generating a representative set that is representative for the data within the set, - inputting directly or indirectly the representative set to a neural network (26) that is trained to retrieve from the input data a measure indicating a proper or improper positioning of the hearing device (2) within the ear canal, and - issuing a notice to a user of the hearing device (2) at least when the measure indicates an improper positioning.