How mayciii/object_scanner works
YOLO Voice Detector reads a phone or webcam stream, detects objects on the PC, announces newly detected or still-visible items, and displays annotated frames. The main loop coordinates inference, announcement policy, display, and keyboard controls; a background stream reader keeps frames current, while a queued speaker avoids blocking detection. The supplied excerpts show these runtime connections, but omit portions of the speaker and stream implementations, so uncertain internal links are left out.
Main entry points: main.py Main loop
Components 9
Video input
- Frame reader
main.py
Detection and display
- Main loop
main.py - YOLO inference
main.py - Annotated display
main.py
Voice announcements
- Announcement policy
main.py - Detection phrasing
main.py - Queued speaker
main.py - Speech synthesis
main.py
Other components
- Phone or webcam
Connections 8
- Phone or webcam → Frame reader: supplies frames
README.md - Main loop → Frame reader: reads frames
main.py - Main loop → YOLO inference: runs inference
main.py - Main loop → Announcement policy: checks detections
main.py - Announcement policy → Detection phrasing: formats counts
main.py - Announcement policy → Queued speaker: queues text
main.py - Queued speaker → Speech synthesis: speaks queued text
main.py - Main loop → Annotated display: shows annotated frame
main.py