Frequency Compression in Hearing Aids Preserving Harmonic Structure

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing frequency compression methods in hearing devices disrupt the harmonic structure of speech sounds, leading to audible artifacts due to the fixed filter bank grid not accounting for the harmonic structure of the signal, making them unsuitable for effective sound perception.

Innovation Solution

A method and device for frequency compression in hearing devices that preserve the phase of the signal by obtaining and impressing amplitude information from source channels onto target channels, allowing the harmonic structure to be maintained, thereby reducing artifacts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If frequency compression is implemented using a fixed filter bank grid, then the compression rule can be applied to shift source channels to target channels, but the harmonic structure of speech sounds is destroyed, leading to audible artifacts

Engineering Contradiction:
Improvefrequency compression capabilityVSAvoidaudible artifacts
Core Design Contradiction:
ProductivityVSObject-generated harmful factors

Solution Approach 1:

The patent separates amplitude information from phase information in the frequency channels. By extracting only amplitude information from source channels and applying it to target channels while preserving the original phase information, the method compresses frequency content without disrupting the harmonic structure that depends on phase relationships.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary representation (amplitude information) that mediates between the source and target channels. This intermediary carries the essential spectral envelope information needed for frequency compression while leaving the phase information (which contains harmonic structure) intact.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If the filter bank grid is fixed, then the system structure is simplified and easier to implement, but it cannot account for the harmonic structure of the signal

Engineering Contradiction:
Improvefilter bank structureVSAvoidharmonic structure information
Core Design Contradiction:
Device complexityVSLoss of information

Solution Approach 1:

The patent extracts only the necessary amplitude information from the frequency channels for compression purposes, leaving the phase information (which contains harmonic structure) intact in the target channels. This selective extraction allows fixed filter bank implementation without losing critical harmonic information.

Inventive Principle:
Principle #2Taking out (Extraction)

3Reliability

If spectral components are shifted from source frequency range to target frequency range, then audibility is improved in dead regions, but the fine structure of the signal is disrupted

Engineering Contradiction:
Improveaudibility in dead regionsVSAvoidfine structure
Core Design Contradiction:
ReliabilityVSStability of the object's composition

Solution Approach 1:

Instead of shifting entire frequency channels (which includes both amplitude and phase information), the patent inverts the approach by transferring only amplitude information to the target channels while keeping the phase information from the original signal. This preserves the fine structure while achieving frequency compression for improved audibility.

Inventive Principle:
Principle #13The other way round (Inversion)

Data Source

PatentEP2622879B1Method and device for frequency compression
Publication Date: 2015.11.11 SIVANTOS PTE LTD
  • EP2622879B1 patent drawingFigure 1
  • EP2622879B1 patent drawingFigure 2~3
  • EP2622879B1 patent drawingFigure 4~5

AI summary

The aim of the invention is to reduce artefacts during frequency compression of an audio signal in a hearing device, in particular a hearing aid. Said aim is achieved by a method in which an amplitude information of a source channel, for example a spectral envelope, is obtained from several frequency channels of the audio signal. An amplitude corresponding to the amplitude information is then applied on a signal in a target channel (15) of the several frequency channels, on which the source channel is represented during frequency compression.