Cochlear Implant Signal Processing for Fine Time Structure

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current speech coding strategies for cochlear implants primarily transmit slow-varying signal envelope information, failing to effectively convey fine time structure cues, which are masked by amplitude modulations, and when they do attempt to transmit fine time structure, amplitude modulations from unresolved harmonics interfere and partially mask this information.

Innovation Solution

A method for processing acoustic signals in cochlear implants that involves generating band pass signals for each electrode, determining a sequence signal based on the channel-specific sampling sequence, filtering the envelope to reduce modulations from unresolved harmonics, and weighting the sequence signal with the filtered envelope to enhance the perception of fine time structure, using techniques such as low pass filtering and psychoacoustic masking models.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If current speech coding strategies transmit slow-varying signal envelope information, then speech understanding is improved, but fine time structure cues are lost

Engineering Contradiction:
Improvefine time structure informationVSAvoidsignal processing complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The audio signal is divided into multiple frequency bands using a filter bank, with each band processed independently to extract both envelope and fine time structure cues. This segmentation allows simultaneous transmission of slow-varying envelope information and fast-varying fine structure cues without loss

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from transmitting only one-dimensional envelope information to transmitting both envelope and fine time structure dimensions. By adding the fine time structure dimension through channel-specific sampling sequences, the system captures both slow and fast temporal variations in the audio signal

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Loss of information

If fine time structure is transmitted using channel-specific sampling sequences, then fine time structure perception is improved, but amplitude modulations from unresolved harmonics interfere and mask this information

Engineering Contradiction:
Improvefine time structure informationVSAvoidamplitude modulations from unresolved harmonics
Core Design Contradiction:
Loss of informationVSObject-affected harmful factors

Solution Approach 1:

The patent extracts and removes unwanted amplitude modulations from the envelope signal using a notch filter tuned to the fundamental frequency. This extraction of harmful components allows the fine time structure information to be transmitted without interference from unresolved harmonic modulations

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent converts the harmful amplitude modulations from unresolved harmonics into useful information by using the fundamental frequency notch filter to identify and remove these modulations. The removal process actually enhances the clarity of the fine time structure cues by eliminating the masking effect

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Data Source

PatentUS8489194B2Enhancing fine time structure transmission for hearing implant system
Publication Date: 2013.07.16 MED EL ELEKTROMEDIZINISCHE GERAETE GMBH
  • US8489194B2 patent drawing
  • US8489194B2 patent drawing
  • US8489194B2 patent drawing

AI summary

A system and method of signal processing for a hearing implant. The hearing implant includes at least one electrode, each electrode associated with a channel specific sampling sequence. An acoustic audio signal is processed to generate for each electrode a band pass signal representing an associated band of audio frequency. For each electrode, a sequence signal is determined as a function of the electrode's associated band pass signal and channel specific sampling sequence. An envelope of each band pass signal is determined. The envelope of each band pass signal is filtered to reduce modulations resulting from unresolved harmonics, creating for each electrode an associated filtered envelope signal. Each electrode's sequence signal is weighted based, at least in part, on the electrode's associated filtered envelope signal.