Mobile Terminal Face Detection Microphone Power Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile terminals consume excessive power due to always-active voice recognition functions, lacking efficient methods to detect the start and end of voice recognition modes.
Innovation Solution
Incorporating a low-power image sensor to detect face images and switch the microphone between inactive and active states based on detected conditions, allowing voice recognition only when necessary and terminating it accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the voice recognition function is always active to recognize user voice at any time, then the voice recognition capability is improved, but the power consumption increases significantly
Solution Approach 1:
The system performs preliminary detection using the low-power image sensor to detect face images before activating the high-power voice recognition function. This preliminary action allows the system to prepare for potential voice recognition needs while consuming minimal power, and only activates the full voice recognition capability when a face is detected, thus resolving the contradiction between always-available voice recognition and power consumption
Solution Approach 2:
The system dynamically switches between different operational states: idle state (image sensor only), voice recognition preparation state (image sensor + microphone), and voice recognition execution state (full voice recognition system). This dynamic state transition allows the system to adapt power consumption to actual usage needs, improving ease of operation only when necessary while minimizing power consumption during idle periods
2Loss of time
If the microphone remains in active state to receive voice information continuously, then the voice recognition response time is improved, but the standby power consumption increases
Solution Approach 1:
The system activates the microphone in advance when a face image is detected, preparing the voice recognition system before actual voice input is received. This preliminary activation reduces the effective response time compared to cold-start scenarios while avoiding continuous activation, as the microphone is only activated when a face is present and deactivated when no face is detected
Solution Approach 2:
The system uses periodic face detection through the low-power image sensor to determine when to activate the microphone. Instead of continuous activation, the microphone is activated periodically based on face detection events, creating a rhythm of activation and deactivation that reduces standby power consumption while maintaining responsive voice recognition when needed
3Productivity
If the voice recognition function is activated frequently based on face detection, then the responsiveness to user needs is improved, but the overall power consumption due to frequent mode switching increases
Solution Approach 1:
The low-power image sensor serves as an intermediary between the idle state and the full voice recognition mode. Instead of directly switching from idle to high-power voice recognition mode, the system uses the image sensor as an intermediate detection layer that triggers microphone activation, creating a staged transition that reduces the frequency and impact of full mode switches while maintaining responsiveness to user needs
Data Source
Figure 1a
Figure 1b~1c
Figure 2~3
AI summary
A mobile terminal including a wireless communication processor configured to provide wireless communication; a microphone; a low-power image sensor configured to detect a subject based on a black and white image sensed by the low-power image sensor; and a controller configured to control the microphone to be in an inactivate state, in response to detecting a face image included in the black and white image sensed by the low-power image sensor, and switch the microphone from the inactive state to an active state and execute a voice recognition function to receive voice information from the subject.