Hearing System Sensor Data Collection via Classifier
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for collecting and storing sensor data for hearing systems are cumbersome, expensive, and laborious, making it difficult to acquire valuable data for fitting, configuring, and training hearing device algorithms and models.
Innovation Solution
A method that involves a hearing device and optionally a mobile device, which uses sensors to detect situations of interest through a classifier implemented in the system, selectively collecting and storing sensor data when a situation of interest is detected, and sending it to a storage system for use in training and configuring hearing devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If sensor data is continuously collected and stored for hearing system training, then the quality and quantity of training data is improved, but the storage requirements and energy consumption increase significantly
Solution Approach 1:
The system performs preliminary classification of sensor data using a classifier before deciding whether to collect and store the data. This preliminary action filters out irrelevant data beforehand, ensuring that only data representing situations of interest are collected, thereby reducing overall storage requirements and energy consumption while maintaining training data quality
Solution Approach 2:
The system applies different data collection strategies based on the local characteristics of each data segment. By classifying each situation and selectively collecting only relevant data, the system optimizes resource allocation locally rather than uniformly collecting all data, reducing total storage and energy usage while preserving essential training data
2Loss of information
If all sensor data is collected for training purposes, then the comprehensiveness of training data is improved, but the processing time and computational resources increase
Solution Approach 1:
The classifier performs preliminary filtering of sensor data to identify situations of interest before the data is processed further. This preliminary classification eliminates irrelevant data early in the pipeline, reducing the total volume of data that requires extensive processing while ensuring that all relevant training information is retained
Solution Approach 2:
The system extracts only the essential and relevant portions of sensor data that represent situations of interest, separating them from irrelevant data. This extraction process maintains the completeness of necessary training information while significantly reducing the processing burden by eliminating unnecessary data
3Measurement precision
If a comprehensive classifier is implemented to accurately detect situations of interest, then the precision of data selection is improved, but the device complexity increases
Solution Approach 1:
The classifier is divided into multiple independent components or modules, each responsible for detecting specific characteristics of situations of interest. This segmentation allows for more precise detection through specialized sub-classifiers while managing overall system complexity by organizing functionality into manageable, modular units
Data Source
AI summary
A method for collecting and storing sensor data (56, 64) of a hearing system (10) comprises: receiving the sensor data (56, 64) of at least one sensor (20, 32, 34) of a hearing device (12) of the hearing system (10), wherein the hearing device (12) is worn by a user; detecting a situation (72) of interest by classifying at least a part of the sensor data (56, 64) with a classifier (61) implemented in the hearing system (10); collecting the sensor data (56, 64), when the hearing system (10) is in a situation (72) of interest; and sending the collected sensor data (76) to a storage system (54) in data communication with the hearing system (10).