Hearing Aid Signal Processing for Conversation Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hearing apparatuses struggle to accurately detect communication situations, such as conversations, due to misclassifications when multiple hearing situations are present, leading to subjectively unfavorable settings and reduced user comfort.
Innovation Solution
A method that involves ascertaining a characteristic voice measure and a supplementary activity measure, evaluating their correlation, and using the result to adjust the signal processing algorithm to enhance the detection of communication situations by increasing the probability value for the presence of a conversation, thereby improving the precision of signal processing in the presence of dominant background noises like music or traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a classifier is used to detect hearing situations based on ambient sounds, then the hearing apparatus can adapt signal processing parameters to different situations, but misclassifications occur when multiple hearing situations are present simultaneously (e.g., music with dominant proportion and voice components with minor proportion)
Solution Approach 1:
The patent segments the detection process into multiple independent measures: a voice measure for detecting voice components, a music measure for detecting music components, and a supplementary measure for detecting wearer activity. Each measure operates independently to evaluate specific aspects of the acoustic environment, allowing the system to detect communication situations more accurately even when multiple hearing situations are present simultaneously.
Solution Approach 2:
The patent introduces an intermediary evaluation process that combines multiple measures (voice measure, music measure, supplementary measure) to determine the probability of a communication situation. This intermediary layer processes the individual measures and integrates them with temporal correlation analysis, serving as a mediator between raw acoustic detection and final classification decisions, thereby improving detection accuracy.
2Device complexity
If the classifier interprets foreground hearing situations as primary and ignores background situations, then the signal processing can be simplified, but this leads to unfavorable settings when the ignored background situation should actually be prioritized (e.g., voice in background while music is in foreground)
Solution Approach 1:
The patent implements feedback through temporal correlation analysis, where the system continuously monitors the relationship between voice measure and supplementary measure over time. This feedback mechanism allows the system to recognize patterns indicative of communication situations and adjust signal processing parameters accordingly, improving user comfort without significantly increasing processing complexity.
Solution Approach 2:
The patent changes the parameter evaluation approach by introducing probability values that combine multiple measures. Instead of relying on a single foreground detection, the system adjusts parameters based on the combined probability derived from voice measure, music measure, and supplementary measure, allowing flexible adaptation to different hearing situations while maintaining manageable complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach increases the precision of detecting communication situations, reduces the need for manual adjustments, and enhances user comfort by providing more accurate and smooth transitions between different signal processing settings, even in complex hearing environments.
Implementation Method 1
a microphone for converting ambient sounds into a microphone signal
Implementation Method 2
an output transducer for outputting the output signal to the ear of a wearer of the hearing apparatus
Implementation Method 3
what is known as a bone conduction receiver
Data Source
AI summary
A hearing apparatus has a microphone for converting ambient sounds into a microphone signal, a signal processor for processing the microphone signal and an output transducer for outputting an output signal. A characteristic voice measure is ascertained for a voice component in the ambient sounds, and a supplementary measure characteristic of an activity of the wearer of the hearing apparatus is ascertained. Subsequently, an evaluated correlation between the voice measure and the supplementary measure is used to increase a probability value for the presence of a communication situation between the user and a third party if the voice measure and the supplementary measure, under at least one prescribed criterion, assume a value representative of the presence of the voice component and of the activity of the wearer. The probability value forms the basis for altering a signal processing algorithm that is executed to process the microphone signal.