Technical Details

OCR Engine

FreezeText uses Apple's Vision framework (VNRecognizeTextRequest) at the highest accuracy level. Text recognition works entirely locally — no data is sent to external servers.

Processing time is typically under 0.3 seconds, depending on the size of the selected area and the amount of text.

Screen Capture

Screen capture uses CoreGraphics (CGDisplayCreateImage). All connected displays are captured simultaneously. The overlay is displayed as a fullscreen window at the highest window level to create the "freeze" effect.

QR Code and Barcode Detection

In addition to text recognition, FreezeText automatically performs QR code and barcode detection in the selected area. Both recognitions run in parallel via the Vision framework.

macOS Services Integration

FreezeText registers as a macOS Service and accepts image files from Finder or other applications. OCR is performed directly on the provided images without requiring a screen capture.

Privacy

• 100% local processing — no cloud connection
• No registration required
• No analytics or tracking data
• OCR via Apple's Vision framework directly on your Mac
• History data stored locally at ~/Library/Application Support/FreezeText/

System Requirements