Environment-Aware Hearing Device Fitting With Preference Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods for fitting and tuning hearing devices do not adequately account for user preferences in different environments, leading to a tedious and inefficient process.
Innovation Solution
A fitting agent that initializes a user model and environment model, allowing for environment-dependent user preference functions, which are updated based on user inputs and environment data to optimize hearing device parameters efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional rule-based methods (NAL-NL1 or NAL-NL2) are used for fitting hearing devices, then the fitting process can be completed with basic audiogram data, but the solution does not account for specific user preferences in different environments
Solution Approach 1:
The patent segments the fitting process into distinct phases: initialization phase using audiogram data with traditional rules, and iterative optimization phase using preference learning. The user model is segmented into multiple components including preference functions, environment models, and parameter distributions that can be updated independently based on user feedback.
Solution Approach 2:
The system transitions from static rule-based fitting to dynamic preference learning. The user model dynamically adapts by updating preference functions and environment models based on real-time user feedback. The fitting process becomes an iterative loop where parameters are continuously refined based on user responses to test settings.
2Measurement precision
If comprehensive user feedback is collected and stored for model updates, then user preference modeling accuracy is improved, but memory requirements increase
Solution Approach 1:
The patent transforms raw user feedback into updated model parameters rather than storing all historical data. The preference functions and environment models are parameterized, allowing the system to represent complex user preferences with a limited set of parameters that can be efficiently stored and processed.
Solution Approach 2:
The system creates simplified representations (copies) of user preferences through parameterized models. Instead of storing actual user responses and environmental data, the system maintains parametric models that replicate user preferences, significantly reducing storage requirements while preserving modeling accuracy.
3Adaptability or versatility
If environment-dependent user preference functions are implemented, then listening experience in various environments is improved, but computational complexity increases
Solution Approach 1:
The patent implements environment-dependent preference functions where different user preferences are modeled for different acoustic environments. The system identifies the current environment and applies the corresponding preference function, allowing tailored optimization for each environment without computing all possible preferences simultaneously.
Solution Approach 2:
The system pre-initializes multiple preference functions for different environments during the setup phase. When operating, the system quickly selects and applies the appropriate pre-initialized function based on environment recognition, avoiding the computational burden of creating and optimizing preference functions in real-time for each environment.
4Measurement precision
If automated preference learning is implemented with multiple test settings, then fitting accuracy is improved, but the fitting process time increases
Solution Approach 1:
The patent implements an iterative feedback loop where the system presents test settings to the user, receives feedback on preferred settings, and uses this feedback to update the user model and generate improved test settings. This structured feedback process systematically improves fitting accuracy while managing time through efficient iteration.
Solution Approach 2:
The system performs a limited number of iterative updates rather than exhaustive optimization. After a predetermined number of iterations or when convergence criteria are met, the fitting process concludes, providing a practical balance between achieving sufficient accuracy and limiting the time investment required from the user and clinician.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A fitting agent for a hearing device system comprising a hearing device worn by a hearing device user is disclosed, wherein the fitting agent comprises one or more processors configured to initialize a user model and an environment model, the user model comprising a plurality of user preference functions and associated user response distribution; obtain environment data indicative of a present environment; determine a first initial environment probability of a first environment and a second initial environment probability of a second environment based on the environment data and the environment model, obtain a test setting comprising a primary test setting and a secondary test setting for the hearing device based on the first initial environment probability and the second initial environment probability; present the test setting to the hearing device user; obtain a user input of a preferred test setting indicative of a preference for either the primary test setting or the secondary test setting; and update the user model for provision of an updated user model based on the preferred test setting and the environment data.