ocrmac

What it is and what it does

ocrmac wraps Apple's native Vision framework to perform optical character recognition directly on macOS, eliminating the need for external OCR engines like Tesseract or EasyOCR. You pass an image file path or PIL image object and receive a list of recognized text snippets, each with a confidence score and bounding box coordinates. The package supports two backends—Vision (available on macOS 10.15+) and LiveText (macOS Sonoma+)—and lets you tune recognition speed versus accuracy or specify language preferences for post-processing.

The main dependencies are Click (for CLI), pillow (for image handling), and pyobjc-framework-Vision (the bridge to Apple's native framework). It's designed for developers on macOS who want fast, built-in OCR without downloading large model files or managing GPU acceleration, though it only works on macOS systems.

Use it for:

• Extract text from screenshots or documents on macOS without external OCR dependencies
• Batch process images to build searchable text indexes with confidence filtering
• Create annotated images showing recognized text regions via PIL or matplotlib output
• Implement language-specific OCR workflows by setting language preferences like 'en-US' or 'zh-Hans'
• Integrate native macOS OCR into Python workflows where speed and native integration matter more than cross-platform support

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Extracts text from images on macOS using Apple's Vision framework, returning recognized text with confidence scores and bounding boxes.

Yes, if you are on macOS and need straightforward OCR without external dependencies. The package is permissively licensed, has low install friction, and leverages native macOS capabilities. The aging maintenance status (218 days since release) is not a blocker—the last commit was recent and the repository is active—but verify that your macOS version meets the 10.15+ requirement and that LiveText features require Sonoma if you need them.

Install

ocrmac on PyPI

pip install ocrmac

uv add ocrmac

poetry add ocrmac

Installing ocrmac

Before you install

Low install friction with a pure-Python wheel and minimal dependencies. The package is aging (218 days since last release) but the repository remains active with recent maintenance work on build configuration and test infrastructure.

License in practice

MIT license permits unrestricted use, modification, and distribution with minimal legal constraints.

Quickstart

pip install ocrmac

from ocrmac import ocrmac
annotations = ocrmac.OCR('test.png').recognize()
print(annotations)  # Returns list of (text, confidence, bounding_box) tuples

Requires macOS 10.15 or later; pyobjc-framework-Vision is a compiled dependency that only installs on macOS.

Verify before relying

• Whether LiveText framework support requires macOS Sonoma or later specifically
• Performance characteristics on Apple Silicon vs Intel Macs beyond the M3 Max benchmark
• Accuracy comparison with the Vision framework's fast vs accurate recognition levels on real-world documents

Package facts

Evidence: ocrmac-1.0.1-py3-none-any.whl

Keywords: ocrmac

Tags

More Text Processing packages