Auditory Prosthesis Noise Reduction via Frequency Domain Gain
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Auditory prosthesis patients face difficulty in hearing in noisy environments due to the inability of conventional noise reduction methods to effectively mitigate ambient noise, which leads to diminished audio signal perception.
Innovation Solution
The method involves dividing an audio signal into frequency domain signals, determining noise reduction gain parameters based on signal-to-noise ratios, and applying noise reduction using current steering and N-of-M stimulation strategies to generate noise-reduced frequency domain signals for enhanced listening experiences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If traditional hearing aid noise reduction circuitry is used to process incoming audio signals, then noise reduction is achieved, but the system becomes cumbersome, expensive, and requires redundant computation
Solution Approach 1:
The patent combines noise reduction functionality directly within the auditory prosthesis system by integrating a noise reduction module that processes frequency domain signals from the cochlear implant's own audio processing channels. This eliminates the need for separate traditional hearing aid noise reduction circuitry, reducing system complexity while maintaining effective noise reduction through shared computational resources and unified signal processing architecture
Solution Approach 2:
The auditory prosthesis system's existing audio signal processing components are made multi-functional by enabling them to perform both cochlear implant stimulation and noise reduction operations. The noise reduction module utilizes the same frequency domain analysis channels already present in the cochlear implant system, allowing these components to serve dual purposes and eliminate redundant computation
2Measurement precision
If noise reduction is applied to improve speech recognition in noisy environments, then speech recognition improves, but computational resources and processing complexity increase
Solution Approach 1:
The noise reduction module applies noise reduction selectively rather than uniformly across all frequency channels. By processing only the frequency domain signals from the cochlear implant's analysis channels and applying noise reduction gain parameters only where needed, the system achieves effective speech recognition improvement while minimizing unnecessary computational energy expenditure
Solution Approach 2:
The system dynamically adjusts noise reduction gain parameters based on the signal-to-noise ratio calculations for each frequency channel. This adaptive parameter adjustment allows the noise reduction algorithm to optimize its computational effort, applying stronger noise reduction only when and where it is most needed, thereby reducing overall computational energy requirements while maintaining speech recognition accuracy
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An exemplary method of reducing an effect of ambient noise within an auditory prosthesis system includes dividing an audio signal presented to an auditory prosthesis patient into a plurality of analysis channels each containing a frequency domain signal representative of a distinct frequency portion of the audio signal, determining a signal-to-noise ratio and a noise reduction gain parameter based on the signal-to-noise ratio for each of the frequency domain signals, applying noise reduction to the frequency domain signals in accordance with the determined noise reduction gain parameters to generate a noise reduced frequency domain signal corresponding to each of the analysis channels, and generating one or more stimulation parameters based on the noise reduced frequency domain signals and in accordance with at least one of a current steering stimulation strategy and an N-of-M stimulation strategy. Corresponding methods and systems are also disclosed.