Smart Home Voice Control Using Face-Gated Command Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional home appliances face issues with voice recognition due to noise interference and misrecognition, especially when users converse near the appliance without intending to give commands, leading to malfunction.
Innovation Solution
The smart home appliance system incorporates a voice input unit, body detection unit, and control units to recognize user intent through voice keywords, facial recognition, and environmental noise filtering, ensuring accurate command recognition and preventing miscommunication among multiple appliances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If voice recognition is activated continuously to capture user commands, then response speed is improved, but noise misrecognition increases
Solution Approach 1:
The system performs preliminary detection of user presence and attention state before activating full voice recognition processing. The control unit detects whether a user is present and paying attention through sensors, and only then activates the voice recognition unit, preventing premature or erroneous recognition while maintaining quick response when ready
Solution Approach 2:
The system uses feedback from multiple sensors (camera, microphone, proximity sensors) to continuously monitor the recognition environment. The control unit adjusts voice recognition activation based on feedback about user presence, attention, and environmental noise levels, dynamically balancing response speed and accuracy
2Ease of operation
If voice recognition is activated continuously to capture user commands, then user convenience is improved, but energy consumption increases
Solution Approach 1:
Instead of continuous voice recognition processing, the system uses periodic detection cycles. The control unit periodically checks for user presence and attention state, and only activates the energy-intensive voice recognition unit when conditions are appropriate, reducing overall energy consumption while maintaining user convenience
Solution Approach 2:
The system performs preliminary low-power detection of user presence and attention before activating full voice recognition. This preliminary check uses less energy-intensive sensors and processing, enabling the system to prepare for voice commands without continuously consuming high energy
3Adaptability or versatility
If multiple home appliances are equipped with voice recognition, then functionality is improved, but miscommunication between appliances increases
Solution Approach 1:
The system uses an intermediary identification process to determine which appliance should receive the voice command. The control unit analyzes the recognition environment and identifies the intended target appliance before executing the command, preventing miscommunication between multiple voice-equipped devices
Data Source
Figure 1~2
Figure 3~4
Figure 5~7
AI summary
Provided is a smart home appliance. The smart home appliance includes: a voice input unit collecting a voice; a voice recognition unit recognizing a text corresponding to the voice collected through the voice input unit; a capturing unit collecting an image for detecting a user's visage; a memory unit mapping the text recognized by the voice recognition unit and a setting function and storing the mapped information, and storing a keyword information that a user may input to start a voice recognition service; a control unit determining whether to perform a voice recognition service on the basis of at least one information of image information collected by the capturing unit and voice information collected by the voice input unit; a region recognition unit determining a user's region on the basis of information on the voice collected through the voice input unit; an output unit outputting region customized information on the basis of information on a region determined by the region recognition unit and information on the setting function, and wherein the control unit comprises a face detection unit recognizing that a user is in a staring state for voice input when image information on a user's visage is collected for more than a setting time through the capturing unit and, wherein the control unit determines that a voice recognition service standby state is entered when it is recognized that there is the keyword information in a voice through the voice input unit and a user is in the staring state through the face detection unit.