Wearable Wake-Word Detection Using Microphone Angle of Arrival
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Healthcare workers face inconvenience in using conventional wireless communication devices due to the need to hold and physically interact with them for operations like powering on, dialing calls, and engaging with graphical interfaces, and 'always-on' devices suffer from significant battery drain, while discontinuous operation poses challenges in detecting user-specific wake words.
Innovation Solution
A lightweight, wearable device equipped with a linear microphone array and low-power voice recognizer detects user-specific wake words to transition between full-power and low-power states, optimizing battery usage and ensuring accurate wake-word detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the device operates in always-on state to enable quick voice commands, then the ease of operation is improved, but the energy consumption increases significantly
Solution Approach 1:
The device dynamically adjusts its operational state based on activity level, transitioning between full-power mode (when active), low-power mode (during idle periods), and sleep mode (when no activity detected). This dynamic state management allows the device to maintain ease of operation when needed while conserving energy during inactive periods.
Solution Approach 2:
The device employs periodic wake-up cycles where the processor briefly activates to check for incoming communications or user input, then returns to sleep mode. This periodic action pattern enables the device to remain responsive to urgent events while minimizing overall energy consumption during extended idle periods.
2Use of energy by moving object
If the device transitions to low-power state to conserve battery, then the energy consumption is reduced, but the ability to detect wake words accurately deteriorates
Solution Approach 1:
The audio processing functionality is segmented into two distinct components: a low-power wake-word detector that remains active during sleep state, and a full-power processor that handles complete audio analysis when activated. This segmentation allows the system to maintain wake-word detection capability with minimal energy consumption while preserving accurate user identification when the full processor engages.
Solution Approach 2:
A dedicated low-power wake-word detection module serves as an intermediary between the sleep state and full-power operation. This intermediary continuously monitors for wake words with minimal energy consumption and acts as a trigger to activate the full processor when a wake word is detected, ensuring reliable user-specific wake word recognition without requiring the main processor to remain constantly active.
3Measurement precision
If the device uses a full processor for wake-word detection, then the wake-word detection accuracy is improved, but the energy consumption increases
Solution Approach 1:
The audio processing functionality is segmented into two distinct components: a low-power wake-word detector that remains active during sleep state, and a full-power processor that handles complete audio analysis when activated. This segmentation allows the system to maintain wake-word detection capability with minimal energy consumption while preserving accurate user identification when the full processor engages.
Solution Approach 2:
Instead of using the full processor capability for all detection tasks, the system applies partial action by employing a simplified low-power detection algorithm sufficient for wake-word recognition. This partial processing approach provides adequate detection accuracy for the specific task while consuming significantly less energy than full processor operation would require.
4Use of energy by moving object
If the device remains in sleep state to conserve power, then the energy consumption is reduced, but the responsiveness to user input deteriorates
Solution Approach 1:
The low-power wake-word detector performs preliminary action by continuously monitoring for wake words even when the main processor is in sleep state. This preliminary detection ensures that when a wake word is spoken, the system can immediately activate the full processor, maintaining fast responsiveness without requiring the main processor to remain constantly active.
Solution Approach 2:
The system implements a feedback mechanism where the low-power detector continuously monitors the audio environment and provides immediate feedback when a wake word is detected. This feedback triggers the transition from sleep state to active state, ensuring rapid response to user input while maintaining energy efficiency during normal operation.
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
Enables hands-free, voice-controlled communication with efficient power management, reducing battery drain and accurately distinguishing user-specific wake words from nearby utterances.
Implementation Method 1
a processor to make a determination, based at least on an angle of arrival of the audio waveform at the at least two microphones of the linear microphone array, of whether to accept the wake-word utterance or rather to reject the wake-word utterance
Data Source
AI summary
Wake-word processing by a wearable electronic device could be carried out when the device is worn by a user and is in a device sleep state, the device including a linear microphone array having at least two microphones vertically spaced from each other, and the device also including a processor. And the example method could involve (i) the at least two microphones of the linear microphone array receiving an audio waveform representing a wake-word utterance, (ii) the processor making a determination, based at least on an angle of arrival of the audio waveform at the at least two microphones of the linear microphone array and/or an energy level of the audio waveform received at the at least two microphones of the linear array, of whether to accept the wake-word utterance or rather to reject the wake-word utterance, and (iii) the processor controlling operation of the device based on the determination.