Wearable Motion Gesture Wake Detection for Speech Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Speech recognition systems face challenges in noisy environments with low signal-to-noise ratios, privacy concerns, and inefficient use of computing resources, particularly when users need to provide commands or feedback in crowded situations.
Innovation Solution
A local wearable device equipped with motion sensors detects user movements to serve as a wake command, allowing users to respond with both spoken and non-verbal gestures, reducing the need for continuous audio transmission and enhancing privacy by enabling motion-based input processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If speech recognition systems continuously process audio inputs, then command recognition capability is maintained, but computing resources are wasted and privacy concerns arise in noisy environments
Solution Approach 1:
The system performs preliminary action by having the wearable device detect wake gestures (hand waves, head movements) before processing audio commands. This preliminary gesture detection activates the speech processing only when needed, preventing continuous resource consumption while maintaining command recognition capability.
Solution Approach 2:
The motion sensor detection acts as an intermediary between the user and the audio processing system. Instead of continuously analyzing audio, the system uses motion sensors as an intermediate trigger to activate audio processing only when a wake gesture is detected, reducing computing resource consumption.
2Ease of operation
If audio processing is used for command recognition, then speech-based interaction is enabled, but privacy concerns increase and processing overhead increases in crowded situations
Solution Approach 1:
The system extracts the wake command function from continuous audio processing and separates it into discrete motion gesture detection. By taking out the activation function from continuous audio analysis and implementing it through motion sensors, the system reduces privacy concerns while maintaining ease of operation.
Solution Approach 2:
The patent replaces acoustic field processing with mechanical motion sensing for wake command detection. Instead of using microphones to detect speech continuously, the system uses motion sensors to detect physical gestures, substituting the acoustic mechanism with a mechanical/optical one that has lower privacy implications.
3Use of energy by moving object
If motion sensors detect wake gestures, then computing resources are conserved, but device complexity increases
Solution Approach 1:
The wearable device leverages its existing multi-functional sensors (accelerometers, gyroscopes already present for fitness tracking) to perform wake gesture detection. By making these existing sensors serve dual purposes (fitness tracking and wake command detection), the system avoids adding significant complexity while achieving resource conservation.
Data Source
AI summary
A system and method for a wearable device capable of detecting a wake gesture for purposes of capturing and forwarding audio data corresponding to a spoken utterance. The device may wake for purposes of capturing utterance audio data in response to a combination of a wake gesture and wakeword. The wake gesture may enable a wakeword detector. The device may also attempt to detect a wakeword utterance in a noisy environment. In response to determining the noisy environment, the device may receive motion data from a motion sensor; determining the motion data corresponds to a wake gesture, and send the audio data corresponding to an utterance to a remote device for processing. The device may also wake based on a combined confidence of wakeword and wake gesture detection.