Motion Analyzer Voice Acquisition Unit Speaker Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current motion analysis systems face challenges in accurately distinguishing between a wearer's voice and others' voices during conversations, particularly due to similar sound pressure levels from distant microphones, leading to incorrect speaker identification.
Innovation Solution
The system employs two microphones positioned differently relative to the wearer's mouth, with one microphone placed farther away and another closer, to differentiate sound pressure ratios and identify the wearer's voice based on non-linguistic information, such as sound pressure volume, and combines this with acceleration sensor data to determine specific motions like nodding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single microphone is used to detect voice, then the device complexity is low, but the speaker identification accuracy deteriorates
Solution Approach 1:
The voice detection function is segmented into multiple microphones positioned at different locations. The first microphone detects the wearer's voice, while the second microphone detects others' voices. This segmentation allows the system to differentiate between speakers based on spatial positioning and sound pressure characteristics, thereby improving speaker identification accuracy without requiring complex processing of a single microphone's signals.
Solution Approach 2:
The system transitions from a single-point voice detection approach to a multi-dimensional spatial detection approach by positioning microphones at different distances and angles relative to the wearer's mouth. This dimensional expansion creates distinct sound pressure patterns for the wearer versus other speakers, enabling more accurate speaker identification through spatial and acoustic differentiation.
2Measurement precision
If motion detection alone is used to identify gestures, then the device complexity is low, but the motion identification accuracy deteriorates due to similar acceleration patterns
Solution Approach 1:
The system merges motion detection data from the acceleration sensor with voice detection data from the microphones to identify gestures. By combining these different types of detection information, the system can distinguish between head movements during conversation and actual gesture motions, significantly improving motion identification accuracy while maintaining relatively simple device architecture.
Solution Approach 2:
The voice detection function serves as an intermediary to disambiguate motion detection results. When the acceleration sensor detects head movement, the system uses voice detection to determine whether the movement is part of normal conversation or a deliberate gesture. This intermediary verification mechanism resolves the ambiguity in motion patterns without requiring complex sensor fusion algorithms.
3Productivity
If voice detection is performed during motion, then the system can identify gestures, but false detection occurs due to head movement affecting microphone input
Solution Approach 1:
The system implements feedback by continuously monitoring both motion and voice detection results and using their correlation to verify gesture identification. When motion is detected, the system checks whether voice detection patterns support the interpretation of a deliberate gesture rather than normal conversation. This feedback mechanism significantly reduces false detections while maintaining high gesture identification productivity.
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 identifies the wearer's voice and detects specific motions with improved accuracy, enhancing the system's ability to analyze conversations and motion analysis in real-time communication scenarios.
Implementation Method 1
a motion detection unit that detects motion of a part of a body of a subject
Implementation Method 2
a speaking detection unit that detects speaking of the subject
Data Source
AI summary
A motion analyzer may include a motion detection unit, a speaking detection unit, and a determination unit. The motion detection unit may detect motion of a part of a body of a subject. The speaking detection unit may detect speaking of the subject. The determination unit may determine that the subject has performed predetermined motion when motion of a part of the body is detected by the motion detection unit and speaking of the subject is detected by the speaking detection unit.