Hearing Aid Sound Processing Using Periodic Aperiodic Signal Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
State-of-the-art hearing aids face challenges in achieving optimal frequency filtering, speech intelligibility, and frequency transposition due to limitations in frequency resolution and temporal smearing, particularly in distinguishing and processing periodic and aperiodic signals.
Innovation Solution
A method and apparatus for processing sound in hearing aids that separate input signals into periodic and aperiodic components, allowing for individual processing and combination to improve frequency filtering, speech intelligibility, and frequency transposition, using a Linear Prediction model and adaptive filters to distinguish voiced and unvoiced sounds, and employing different frequency filter banks for each signal type.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single filter bank is used for all frequency bands, then the device complexity is reduced, but the frequency resolution and speech intelligibility deteriorate
Solution Approach 1:
The patent divides the input signal into periodic and aperiodic components using Linear Prediction, then applies separate filter banks to each component. This segmentation allows different filter configurations optimized for each signal type, improving frequency resolution and speech intelligibility without requiring a single overly complex filter bank for all signals.
2Measurement precision
If high frequency resolution is achieved through traditional filter banks, then speech analysis is improved, but temporal smearing increases
Solution Approach 1:
By separating periodic and aperiodic signals, the patent can apply different processing strategies. The periodic signal component benefits from high frequency resolution filtering, while the aperiodic component is processed to minimize temporal smearing, thus resolving the trade-off between frequency resolution and temporal fidelity.
Solution Approach 2:
Different filter characteristics are applied locally to different signal components. The periodic signal path uses filters optimized for frequency resolution, while the aperiodic signal path uses filters optimized for temporal response, allowing each component to be processed with the most appropriate characteristics for its nature.
3Device complexity
If uniform processing is applied to all sound signals, then the device complexity is reduced, but the speech intelligibility and noise reduction capabilities deteriorate
Solution Approach 1:
The patent segments the signal processing into distinct paths for periodic and aperiodic components, allowing specialized processing for each. This enables tailored filtering, amplification, and noise reduction strategies that improve speech intelligibility without requiring a single overly complex uniform processing structure.
Solution Approach 2:
Different processing parameters and algorithms are applied locally to different signal components based on their characteristics. Voiced sounds receive processing optimized for preserving speech intelligibility, while unvoiced sounds receive processing optimized for their specific characteristics, improving overall system performance without uniform complexity.
Data Source
AI summary
A method of processing sound in a hearing aid (100, 300, 400) comprises separating the input transducer signal into a periodic and aperiodic signal for further processing in the hearing aid (100, 300, 400). The invention also provides a hearing aid (100, 300, 400) adapted for carrying out such a method of sound processing.