Voice Activated Device Proximity Detection for False Activation Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Voice-activated devices often activate unintentionally due to ambient speech, leading to unwanted actions and erroneous command processing due to lower fidelity voice recognition from far-field microphones, which consume system resources and power unnecessarily.
Innovation Solution
Implementing a system that determines user proximity using sensors like luminosity, passive infrared, cameras, and Bluetooth connectivity to enable voice commands only when the user is within a specified radius, thereby reducing false activations and improving voice recognition fidelity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If voice activation is enabled continuously, then the device responds to all voice commands, but it activates unintentionally due to ambient speech and consumes excessive system resources and power
Solution Approach 1:
The system performs preliminary proximity detection using sensors (luminosity, infrared, camera, Bluetooth) before enabling voice recognition. This preliminary action determines whether the user is within the specified radius, and only if proximity is confirmed does the system activate voice processing, thereby preventing unnecessary power consumption while maintaining reliable voice command accuracy when needed
Solution Approach 2:
The voice activation state is made dynamic rather than static. The system continuously monitors proximity data from multiple sensors and dynamically adjusts the voice activation state based on real-time conditions. When the user enters the proximity zone, voice features are enabled; when leaving, they are disabled. This dynamic approach ensures power is consumed only when necessary while maintaining high reliability of voice recognition
2Reliability
If voice recognition processes are always active, then all voice commands are processed, but erroneous commands are processed due to lower fidelity from far-field microphones
Solution Approach 1:
Before voice recognition processing occurs, the system performs preliminary verification by checking proximity data from sensors. This preliminary action confirms that the user is within the specified radius of the device, ensuring that only high-fidelity near-field voice commands are processed. This prevents erroneous processing of far-field ambient speech while maintaining high productivity by avoiding unnecessary processing of invalid commands
Solution Approach 2:
The proximity detection system acts as an intermediary between the user and the voice recognition process. Multiple sensors (luminosity, infrared, camera, Bluetooth) serve as intermediary detectors that verify user presence before allowing voice commands to be processed. This intermediary layer ensures that only authenticated, proximate users can activate voice features, thereby maintaining high recognition fidelity and processing efficiency
3Reliability
If proximity detection using multiple sensors is implemented, then false activations are reduced, but device complexity increases
Solution Approach 1:
The proximity detection function is segmented across multiple independent sensors rather than using a single complex sensor system. Each sensor (luminosity, infrared, camera, Bluetooth) independently contributes to proximity determination. This segmentation allows the system to achieve high activation accuracy through data fusion from multiple simple, well-understood sensor types, managing complexity by dividing the detection function across independent components
Solution Approach 2:
Multiple sensors with different primary functions are utilized for proximity detection. The luminosity sensor, infrared sensor, camera, and Bluetooth module are multi-functional components that serve their primary purposes while also contributing to proximity determination. This universality approach reduces overall system complexity by repurposing existing multi-functional components rather than adding dedicated proximity sensors
Data Source
AI summary
Techniques for controlling a voice activated feature of a voice activated device are described. Data from one or more sensors and data indicative of a status of a user are received. Based on the analyzing the data, a proximity of the user relative to the location of the voice activated device is determined. One or more voice activated features of the voice activated device are enabled based at least in part on the determined proximity, one or more rules, and one or more user preferences.