Hearing Aid Preference Elicitation With Cloud-Based Personalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hearing aid personalization methods are challenging due to difficulty in user preference elicitation, high processing and memory requirements, and the need for complex user interaction, making it difficult to adapt settings to varying user preferences efficiently.
Innovation Solution
A hearing aid system with a portable computer device and interactive display that uses machine learning procedure screens to optimize settings based on user input, time budget, listening intent, and sound environment, allowing for real-time personalization with limited user interaction and processing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If Bayesian incremental preference elicitation is used to optimize hearing aid parameters, then personalization can be achieved, but the processing and memory requirements become very high
Solution Approach 1:
The patent extracts the preference modeling function from the hearing aid processor and implements it in a cloud-based server. This separates the computationally intensive Bayesian preference elicitation from the hearing aid device, allowing the device to have limited processing and memory resources while still achieving personalized optimization through cloud computing support.
Solution Approach 2:
The patent introduces a cloud-based server as an intermediary between the hearing aid device and the user's hearing preferences. This intermediary handles the complex Bayesian inference and preference modeling, enabling the hearing aid to achieve personalization without requiring high processing power and memory capacity built into the device itself.
2Adaptability or versatility
If parameterized approach is used to model user response function, then personalization can be achieved, but the method becomes complicated and difficult to implement
Solution Approach 1:
The patent extracts the complex parameterized preference modeling from the hearing aid device and relocates it to a cloud-based server. This allows the device to maintain simplicity while the server handles the complicated Bayesian inference and preference function modeling, making the overall system more manageable and easier to implement.
3Measurement precision
If extensive user interaction is required for personalization, then optimization can be achieved, but the time required and user burden increase
Solution Approach 1:
The patent implements self-service preference elicitation where the system automatically captures user feedback through simple rating mechanisms (e.g., thumbs up/down) without requiring extensive manual input. The Bayesian inference system automatically processes these simple responses to model user preferences, significantly reducing the time and effort required from the user while maintaining accurate optimization.
Solution Approach 2:
The patent incorporates feedback mechanisms where users provide simple ratings of hearing aid performance, and the system uses this feedback to iteratively refine the preference model through Bayesian inference. This feedback loop enables continuous optimization with minimal user input, reducing the overall time required for personalization while maintaining high accuracy.
Data Source
AI summary
A method (100, 200) of optimizing a hearing aid system.