Accelerometer-Based Voice Activity Detection in Head-Worn Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional voice activity detection systems using microphones are prone to errors due to ambient noise, such as wind and other environmental sounds, which can lead to false positives and negatives in detecting user voice commands.
Innovation Solution
A head-worn electronic device equipped with an accelerometer having multiple axes to detect three-dimensional forces, which processes vibration vectors to determine a voice activity detection axis that correlates with user voice vibrations, allowing for accurate voice activity detection and control of host devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If microphone-based VAD systems are used, then voice detection capability is provided, but detection accuracy deteriorates due to ambient noise interference
Solution Approach 1:
The patent replaces the acoustic field-based microphone detection system with a mechanical field-based accelerometer detection system. The accelerometer detects vibrations caused by voice activity through bone conduction and tissue transmission, which are mechanical vibrations that propagate through the user's head and ear structures. This substitution eliminates the microphone's vulnerability to ambient acoustic noise while maintaining voice detection capability through a different physical domain (mechanical vibrations instead of acoustic waves).
2Object-affected harmful factors
If accelerometer-based detection is implemented, then resistance to ambient noise is improved, but device complexity increases due to additional processing requirements
Solution Approach 1:
The patent applies local quality by focusing processing efforts on specific characteristics of the acceleration signal that are unique to voice activity. Instead of processing the entire three-dimensional acceleration vector equally, the system identifies and processes only the specific vibration patterns and frequency ranges that correspond to voice-induced mechanical vibrations in the ear. This selective processing approach reduces computational complexity while maintaining high noise resistance by concentrating resources on the most informative signal features.
3Measurement precision
If three-dimensional vibration vectors are processed, then voice detection precision is improved, but computational requirements increase
Solution Approach 1:
The patent extracts and processes only the essential components of the three-dimensional vibration vector that are relevant to voice activity detection. By analyzing the acceleration data along different axes and identifying the specific directional patterns characteristic of voice-induced vibrations, the system extracts the meaningful signal components while discarding redundant information. This extraction approach maintains high detection precision by focusing on the most informative dimensions of the vibration data while reducing overall computational energy consumption compared to exhaustive processing of all possible signal characteristics.
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
The solution improves voice activity detection accuracy by minimizing false positives and negatives, providing a better user experience and extending battery life by leveraging accelerometer-based vibration detection that is less affected by ambient noise.
Implementation Method 1
an accelerometer having a plurality of axes for detecting three-dimensional forces applied to the head worn electronic device
Data Source
AI summary
An example embodiment includes a head worn electronic device comprising a transceiver for communicating with a host device, an accelerometer having a plurality of axes for detecting three-dimensional forces applied to the head worn electronic device, and a processor. The processor is configured to receive a three-dimensional vibration vector from the accelerometer caused by a voice of a user while the head worn electronic device is positioned in a user's ear, process the three-dimensional vibration vector to determine a voice activity detection axis that correlates with vibrations caused by the voice of the user, perform processing of data from the voice activity detection axis to detect voice activity of the user, and send an instruction to the host device via the transceiver to control the host device based on the voice activity detection.