Auto Audio Video Focusing via Microphone Array Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current electronic devices are limited in providing satisfactory video capture by failing to automatically focus on multiple objects of interest and require separate user inputs for audio and video focusing, leading to user dissatisfaction.
Innovation Solution
An electronic device equipped with a capturing unit, microphone array, and processor that registers objects of interest, infers the primary object, adjusts video zoom, and performs audio focusing by adjusting microphone activity based on the object's position, eliminating the need for separate user inputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automatic audio and video focusing is implemented, then user satisfaction and capture quality are improved, but device complexity increases due to multiple microphones and processing requirements
Solution Approach 1:
The system segments the audio capture function by using multiple microphones with different directivity patterns (omnidirectional, supercardioid, cardioid) to capture sound from different directions. Each microphone captures specific spatial information, and the processor segments and combines these signals to achieve focused audio capture on selected objects while suppressing background noise.
Solution Approach 2:
The microphone array system performs multiple functions: it captures audio from all directions simultaneously, identifies multiple speaking objects, performs audio focusing on selected objects, and suppresses background noise. The same hardware infrastructure supports both audio focusing and video zooming operations.
2Measurement precision
If separate user inputs are required for audio and video focusing, then focusing precision is maintained, but ease of operation deteriorates due to frequent user inputs
Solution Approach 1:
The system performs automatic audio focusing and video zooming without requiring user input. The processor automatically identifies speaking objects using audio signals, determines their positions, and controls the capturing units to focus on these objects. The system serves itself by autonomously completing the focusing task that would otherwise require manual user operation.
Solution Approach 2:
The system uses audio signals as feedback to automatically adjust video focusing. The processor continuously monitors audio inputs, identifies speaking objects, and uses this information to control the capturing unit's zoom and focus, creating a closed-loop system where audio feedback drives video adjustment without user intervention.
3Reliability
If audio focusing is performed on multiple objects of interest, then audio quality is improved, but device complexity increases due to multiple microphones and processing requirements
Solution Approach 1:
The system dynamically adjusts the directivity patterns of microphones based on the number and position of selected speaking objects. When multiple objects are selected, the system dynamically changes microphone activities to focus on different directions simultaneously, adapting the audio capture pattern in real-time to match the spatial distribution of speaking objects.
Data Source
AI summary
Disclosed are a device and a method for providing automatically focusing on audio and video. An operation method of an electronic device comprises registering objects of interest, capturing video, displaying the video on a display unit, inferring at least one object of interest included in the video among the objects of interest, adjusting a zoom of the video being captured by controlling the capturing unit based on a position of the at least one object of interest and performing audio focusing by adjusting activity of each of the multiple microphones based on the position of the at least one object of interest. Accordingly, in capturing video by means of the electronic device, the satisfaction with capturing results can be enhanced by emphasizing the audio of interest and by video focusing.