Adaptive Hearing Aid Acoustic Classification System

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing hearing aids lack adaptability to individual user environments, relying on fixed acoustic classes that may not be relevant and cannot be easily modified or adapted for different users, leading to suboptimal listening experiences.

Innovation Solution

An adaptive environmental classification system that continuously learns and redefines classes based on new environments encountered by the user, using a combination of supervised and unsupervised learning algorithms to split, merge, and modify classes dynamically, allowing for personalized parameter adjustments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If fixed acoustic classes are used for environmental classification, then the system structure is simple and easy to implement, but the adaptability to individual user environments is poor

Engineering Contradiction:
Improveadaptability to individual user environmentsVSAvoidsystem structure complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic class creation and modification capabilities where the classification system can automatically create new acoustic classes, split existing classes, and merge classes based on the user's encountered environments. This transforms the static fixed-class system into a dynamic adaptive system that evolves with user experience, resolving the contradiction between adaptability and system complexity.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The hearing aid system performs self-learning and self-adjustment by automatically analyzing acoustic environments and modifying its classification structure without requiring external intervention. The system serves itself by autonomously adapting to individual user needs, eliminating the need for manual programming while maintaining reasonable complexity through automated processes.

Inventive Principle:
Principle #25Self-service

2Adaptability or versatility

If predefined classes are used to guide the learning process, then the initial classification accuracy is improved, but the scope for adapting the system after training is limited

Engineering Contradiction:
Improvescope for adapting the system after trainingVSAvoidinitial classification accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent employs predefined acoustic classes as an initial starting point for the learning process, providing a head start for classification accuracy. These preliminary classes guide the initial learning phase, after which the system transitions to automatic class creation and modification, thus maintaining both initial accuracy and long-term adaptability.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically transitions from static predefined classes to dynamically created and modified classes. The classification structure evolves from a fixed initial state to a flexible adaptive state, allowing the system to maintain initial classification accuracy while gaining extensive post-training adaptability through automatic class management.

Inventive Principle:
Principle #15Dynamics

3Ease of operation

If the hearing aid changes algorithms for different environments, then the user satisfaction is significantly increased, but the requirement for user interaction through mode switching is added

Engineering Contradiction:
Improveautomatic environmental adaptationVSAvoidcontrol system complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The hearing aid system automatically detects acoustic environments and adjusts its parameters without requiring user intervention. The system serves itself by autonomously performing environmental classification and parameter adjustment, eliminating the need for manual mode switching while providing enhanced user satisfaction through automatic adaptation.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system implements a feedback mechanism where acoustic environment detection results automatically trigger parameter adjustments. The classification output feeds back to the parameter control system, creating a closed-loop automatic control system that adapts to environments without user input, thereby improving ease of operation while managing complexity through automated feedback processing.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP2163124B1Fully learning classification system and method for hearing aids
Publication Date: 2017.08.23 UNIVERSITY OF OTTAWA
  • EP2163124B1 patent drawingFigure 1~2
  • EP2163124B1 patent drawingFigure 3~4
  • EP2163124B1 patent drawingFigure 5

AI summary

A method for operating a hearing aid in a hearing aid system where the hearing aid is continuously learnable for the particular user. A sound environment classification system is provided for tracking and defining sound environment classes relevant to the user. In an ongoing learning process, the classes are redefined based on new environments to which the hearing aid is subjected by the user.