Hearing Device Head Movement Detection via Orthogonal Acceleration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hearing devices struggle to accurately identify conversation participants in noisy environments with multiple speakers, as existing voice recognition algorithms often misinterpret head movements, leading to impaired speech comprehension.
Innovation Solution
A method utilizing a single acceleration sensor positioned on the hearing device wearer's head, configured for two orthogonal measurement axes, to detect yaw movements by analyzing tangential and radial acceleration signals, incorporating criteria such as time characteristics, local extremes, and correlation coefficients to enhance the accuracy of head movement detection and reduce misinterpretation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If voice recognition algorithms are used to detect conversation situations, then speech comprehension is improved, but accuracy deteriorates in situations with multiple speakers
Solution Approach 1:
The patent introduces an intermediary system consisting of multiple microphones and signal processing algorithms that analyze acoustic characteristics to distinguish between different speech sources. This intermediary layer between the raw audio input and the voice recognition algorithm enables accurate identification of conversation participants even when multiple speakers are present, resolving the contradiction between improved speech comprehension and maintained detection accuracy.
2Measurement precision
If multiple sensors are used to detect head movements, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent employs a single acceleration sensor that performs multiple functions: detecting head movements, determining orientation, and providing temporal information about speech sources. By making this single sensor multi-functional and combining its data with acoustic information from multiple microphones, the system achieves high measurement precision without increasing device complexity through multiple sensors.
Solution Approach 2:
The patent compensates for the limitation of a single sensor by adding temporal and acoustic dimensions to the analysis. Instead of relying solely on spatial information from multiple sensors, the system uses the time characteristics of acceleration signals and combines them with acoustic data from multiple microphones, achieving high precision through multi-dimensional analysis rather than multiple spatial sensors.
3Reliability
If complex signal processing is used to identify conversation participants, then speech comprehension is improved, but power consumption increases
Solution Approach 1:
The patent implements a hierarchical signal processing approach where simpler preprocessing steps (such as basic acceleration signal analysis and acoustic feature extraction) are performed continuously, while more complex processing is applied selectively only when needed to resolve ambiguous situations. This partial application of complex processing maintains speech comprehension while reducing overall power consumption compared to continuously applying full complex signal processing.
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 effectively reduces misinterpretations of head movements, allowing for improved identification of conversation participants and enhanced speech comprehension in complex acoustic situations, while minimizing power consumption by eliminating the need for multiple sensors.
Implementation Method 1
an acceleration sensor that is positioned on the head of a hearing device wearer in the intended worn state and that is configured for measurement in two mutually orthogonal measurement axes
Data Source
AI summary
A hearing device has an acceleration sensor that is positioned on the head of a hearing device wearer in the intended worn state, is configured for measurement in two mutually orthogonal measurement axes and is operated by virtue of at least one main feature related to an acceleration directed tangentially in relation to the head being derived from an acceleration signal of the acceleration sensor. The at least one main feature is used to ascertain a presence of a yaw movement of the head by taking into consideration at least one prescribed criterion, derivable from the acceleration signal itself, beyond the presence of an acceleration value of the tangentially directed acceleration that is indicative of a movement.