Hearing Device Sound Signal Modelling for Personalized Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Noise reduction methods in hearing aids often fail to accurately distinguish between noise and target signals, leading to ineffective noise suppression and enhancement, as they rely on assumptions that may not hold in real-world conditions.
Innovation Solution
A hearing device and method that allow users to record specific sound signals, determining parameter values for a sound signal model, which can be applied to process incoming signals for noise suppression or enhancement, enabling personalized noise reduction and target enhancement based on user-recorded object signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional noise reduction methods with prior assumptions are used, then the processing is simple and fast, but the noise suppression effectiveness deteriorates when assumptions do not hold in real-world conditions
Solution Approach 1:
The system performs preliminary recording of target signals and noise signals in advance to build signal models before actual hearing aid operation. This preliminary action creates reusable signal models that improve noise suppression effectiveness without requiring complex real-time processing during actual use.
Solution Approach 2:
The system creates copies of target signals and noise signals through recording, then uses these copies to build signal models. The signal models are derived from recorded signal copies rather than requiring complex real-time analysis of actual mixed signals, thereby improving reliability while maintaining processing simplicity.
2Measurement precision
If user-initiated recording and personalized signal modeling is implemented, then noise reduction accuracy improves, but the ease of operation deteriorates due to additional user actions required
Solution Approach 1:
The system enables users to self-record target signals and noise signals using their own devices. Users independently capture the signals they wish to protect or suppress, and the system automatically processes these recordings to create personalized signal models, improving accuracy while maintaining ease of use.
Solution Approach 2:
The system utilizes existing universal recording devices (smartphones, computers, voice recorders) that users already possess, rather than requiring specialized hearing aid recording functionality. This multi-functional approach leverages common devices for signal capture, improving signal discrimination accuracy without complicating user operations.
3Adaptability or versatility
If signal models are created from recorded object signals, then the adaptability to specific environments improves, but the loss of time increases due to recording and model determination steps
Solution Approach 1:
The system performs signal recording and model determination as preliminary actions before actual hearing aid deployment. By completing these time-consuming setup steps in advance, the system achieves environment-specific adaptability while ensuring that the actual hearing aid operation requires minimal processing time.
Solution Approach 2:
The system creates reusable signal model copies from recorded signals, allowing the same model to be applied repeatedly without requiring repeated recording or processing. This copying approach enables environment-specific adaptation while minimizing time loss during actual use, as the models are determined once and then efficiently reused.
Data Source
AI summary
A hearing device configured to be worn by a user, includes: a first input transducer for providing an input signal; a first processing unit configured for processing the input signal according to a first sound signal model; and an acoustic output transducer coupled to an output of the first processing unit for conversion of an output signal from the first processing unit into an audio output signal; wherein the hearing device is configured to obtain an input signal comprising a first signal part and a second signal part, the first signal part corresponding at least partly to a first object signal recorded by a recording unit; and wherein the hearing device is also configured to apply a first set of parameter values of a second sound signal model to the first sound signal model, and process the input signal according to the first sound signal model.