EasyOCR

An open-source Python OCR library extracts text from photos and documents, supports multilingual images, and uses the Apache 2.0 license.

EasyOCR is a Python library for developers adding text recognition to an application. It reads both text in photographs and dense document text.

Multilingual recognition: Read scripts including Latin, Chinese, Arabic, Devanagari and Cyrillic, with languages such as Japanese, Korean, Hindi and Ukrainian.
Text locations and confidence: Results include the recognized text, its bounding box and a confidence score, so applications can locate text within the original image.
Custom models: Developers can train or use their own recognition models and work with CRAFT detection models.

Language combinations have limits. Languages that share a writing script can work together, and English can accompany any supported language.

EasyOCR uses PyTorch, supports Windows and is open source under Apache 2.0. A separate hosted demo runs on Hugging Face Spaces using Gradio. Jaided AI also offers commercial support for custom OCR training, fine-tuning and deployment.

OCR and Document Scanning Inference Libraries and Bindings Windows Multilingual