Adaptive Filter Bank for Hearing Aid Noise Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional hearing aid systems face challenges in achieving an optimal trade-off between time and frequency resolution for noise suppression and speech enhancement, leading to issues such as echo effects and smearing of transient events, which affect speech intelligibility.

Innovation Solution

The method employs an adaptive filter bank with an aggregate window that grows based on the Likelihood Ratio Test, allowing for variable time and frequency resolution in each frequency bin, using a combination of Hann windows and zero-padding to maintain constant frequency bin count, and applying a Discrete Fourier Transform for efficient processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a long analysis window is used to improve frequency resolution and noise suppression, then frequency resolution is improved, but transient events are smeared and time resolution deteriorates

Engineering Contradiction:
Improvefrequency resolutionVSAvoidtime resolution
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies dynamics by making the analysis window length variable rather than fixed. The system dynamically adjusts the window length based on signal characteristics detected through the Likelihood Ratio Test, allowing long windows for stationary signals (improving frequency resolution) and short windows for transient signals (preserving time resolution). This resolves the contradiction by adapting the window length to match the temporal characteristics of the signal being processed.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of analysis window length from a static value to a dynamically adjustable parameter. By using the Likelihood Ratio Test to detect signal non-stationarity and adjusting the window length accordingly, the system optimizes the trade-off between frequency resolution and time resolution. This parameter change allows the system to achieve high frequency resolution when needed while preserving transient events when required.

Inventive Principle:
Principle #35Parameter changes

2Loss of time

If a short analysis window is used to preserve transient events and improve time resolution, then time resolution is improved, but frequency resolution and noise suppression capability deteriorate

Engineering Contradiction:
Improvetime resolutionVSAvoidfrequency resolution
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The system dynamically selects short analysis windows when transient events are detected through the Likelihood Ratio Test. This dynamic adaptation ensures that transient events like plosives are preserved with high time resolution while avoiding the frequency resolution degradation that would occur with consistently short windows. The window length is adjusted in real-time based on signal characteristics.

Inventive Principle:
Principle #15Dynamics

3Device complexity

If a fixed window length is used to simplify processing, then device complexity is reduced, but speech intelligibility deteriorates due to echo effects and transient smearing

Engineering Contradiction:
Improveprocessing complexityVSAvoidspeech intelligibility
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent changes the processing approach from using a fixed window length to using variable window lengths based on signal analysis. The Likelihood Ratio Test provides a computationally efficient method to detect signal non-stationarity, enabling the system to adapt window lengths without excessive complexity. This parameter change improves speech intelligibility by preventing echo effects and transient smearing while maintaining manageable processing requirements.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system implements feedback through the Likelihood Ratio Test, which continuously analyzes the signal and provides information about its stationary or non-stationary nature. This feedback loop enables the system to adjust the analysis window length in real-time, improving speech intelligibility by adapting to changing signal conditions while maintaining efficient processing through the use of a well-defined decision criterion.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3074975B1Method of operating a hearing aid system and a hearing aid system
Publication Date: 2018.05.09 WIDEX AS
  • EP3074975B1 patent drawingFigure 1
  • EP3074975B1 patent drawingFigure 2
  • EP3074975B1 patent drawingFigure 3

AI summary

A method of operating a hearing aid system using adaptive time-frequency analysis in order to provide improved noise reduction and enhanced speech intelligibility and a hearing aid system (100, 200 ) comprising an adaptive filter bank.