Probabilistic Hearing Aid Compensation for Loudness Restoration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current hearing aids for sensorineural hearing loss lack a proper mathematical description of hearing loss, making it difficult to evaluate and compare different compensation methods, relying heavily on subjective testing and empirical rules.
Innovation Solution
A new hearing aid that performs probabilistic hearing loss compensation using a predetermined hearing loss model, employing a Kalman filtering principle and Bayesian inference to process audio signals and restore loudness, allowing for automatic gain determination and evaluation of hearing loss compensation algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dynamic range compressors are configured based on empirical rules and subjective testing, then hearing loss compensation can be achieved, but objective evaluation and comparison of different compensation methods become difficult
Solution Approach 1:
The patent introduces a probabilistic hearing loss model as an intermediary between the audio signal and the compression algorithm. This model serves as a mathematical mediator that transforms subjective hearing loss characteristics into objective probability distributions, enabling quantitative evaluation while maintaining the empirical basis of hearing aid fitting
Solution Approach 2:
The patent transforms the traditional fixed compression ratio parameter into a probabilistic parameter that varies according to the hearing loss model. By changing from deterministic parameters to probabilistic parameters, the system enables objective evaluation of different compensation strategies while adapting to individual hearing losses
2Productivity
If compression ratio and time constants are adjusted according to empirical fitting rules, then individual hearing loss compensation is achieved, but the lack of mathematical description prevents algorithm optimization
Solution Approach 1:
The patent replaces the mechanical trial-and-error adjustment process with a probabilistic mathematical framework. By substituting empirical rule-based adjustment with probability distribution-based optimization, the system enables algorithmic optimization while preserving the individualized compensation approach
Solution Approach 2:
The patent implements a feedback mechanism where the probabilistic hearing loss model continuously informs the compression algorithm adjustments. The model provides mathematical feedback about the expected perceptual effect of compression parameters, enabling systematic optimization rather than random empirical adjustment
3Adaptability or versatility
If multiband compressors process different frequency bands separately with different parameters, then frequency dependence of hearing loss is accounted for, but the number of parameters to be adjusted increases significantly
Solution Approach 1:
The patent creates a universal probabilistic hearing loss model that can be applied across all frequency bands with a single set of probability distributions. This universal model serves multiple functions: it characterizes hearing loss across frequencies, guides multiband compression parameters, and enables objective evaluation, thereby reducing the effective number of independent parameters to adjust
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A hearing aid is provided, comprising an input transducer for provision of an audio signal in response to sound; a hearing loss model for calculation of a hearing loss as a function of a signal level of the audio signal; and a probabilistic hearing loss compensator that is configured to process the audio signal into a hearing loss compensated audio signal in such a way that the hearing loss is restored to normal hearing in accordance with the hearing loss model.