Hearing Aid Transient Noise Attenuation via LPC Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Hearing assistance devices amplify sounds but also make sharp transient noises annoying, and previous solutions fail to differentiate between annoying environmental transients and important speech transients, leading to reduced audibility of speech.

Innovation Solution

The method involves using linear predictive coding (LPC) to isolate speech and non-speech segments in audio signals, with dynamic threshold calculation and attenuation gain settings based on prediction error magnitude and threshold values to reduce annoyance of non-speech transients while maintaining audibility of speech transients.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If onset detection and attenuation is applied to reduce annoyance of sharp transients, then transient noise annoyance is reduced, but audibility of perceptually important transients in speech is reduced

Engineering Contradiction:
Improvetransient noise annoyanceVSAvoidaudibility of speech transients
Core Design Contradiction:
Object-affected harmful factorsVSReliability

Solution Approach 1:

The patent segments transients into two distinct categories: annoyance transients (environmental noises) and speech transients (perceptually important sounds). By segmenting the transient processing into separate handling paths for each category, the system can attenuate annoyance transients while preserving speech transients, thus resolving the contradiction between reducing noise annoyance and maintaining speech audibility

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different processing qualities to different transient types. Annoyance transients receive strong attenuation (high processing intensity), while speech transients receive minimal or no attenuation (low processing intensity). This local differentiation of processing quality allows simultaneous optimization of both noise reduction and speech preservation

Inventive Principle:
Principle #3Local quality

2Object-affected harmful factors

If previous transient attenuation solutions are used, then transient noise is reduced, but discrimination between annoying environmental transients and speech-related transients is poor

Engineering Contradiction:
Improveenvironmental transient noiseVSAvoidtransient classification accuracy
Core Design Contradiction:
Object-affected harmful factorsVSMeasurement precision

Solution Approach 1:

The patent employs dynamic threshold calculation that adapts to the current audio signal characteristics. The threshold is continuously adjusted based on the signal's statistical properties, enabling the system to dynamically distinguish between transient types in real-time. This dynamic adaptation significantly improves classification accuracy compared to static threshold methods

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system uses feedback from the audio signal analysis to continuously refine transient detection and classification. By monitoring the signal characteristics and adjusting the dynamic threshold based on feedback information, the system achieves more accurate discrimination between annoyance and speech transients

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3223278B1Noise characterization and attenuation using linear predictive coding
Publication Date: 2021.04.14 STARKEY LABORATORIES INC
  • EP3223278B1 patent drawingFigure 1
  • EP3223278B1 patent drawingFigure 2~3
  • EP3223278B1 patent drawingFigure 4

AI summary

Disclosed herein, among other things, are apparatus and methods for noise characterization and attenuation for hearing assistance devices. In various embodiments, a method of operating a hearing assistance device includes receiving an audio signal using a microphone of the hearing assistance device and identifying a transient in the audio signal. Linear predictive coding (LPC) is used to isolate speech segments and non-speech segments of the transient and fluctuating noise, and the non-speech segments of the transient and fluctuating noise are attenuated to reduce annoyance of the noise and maintain audibility of perceptually important transients in speech.