Umi-OCR is a free, open source Optical Character Recognition (OCR) application originally developed for Windows by hiroi-sora on GitHub. Since its release, it has grown into one of the most popular open source OCR tools, especially celebrated for its exceptional Chinese text recognition accuracy.
What Makes Umi-OCR Special?
Unlike most OCR tools that depend on cloud APIs or require a subscription, Umi-OCR runs entirely offline. This means:
- Privacy first — your documents never leave your machine
- No API keys or rate limits — process as many documents as you need
- High accuracy — particularly for Chinese, Japanese, and Korean text
- Batch processing — handle hundreds of images in one session
Supported Languages
Umi-OCR supports a wide range of recognition languages including:
- Simplified Chinese (极佳 accuracy)
- Traditional Chinese
- English
- Japanese
- Korean
- Mixed Chinese-English documents
How Umiocr Builds on Umi-OCR
Umiocr takes the Umi-OCR recognition engine and brings it to the web. Instead of installing desktop software, users can now access the same OCR capability from any browser on any device — including smartphones, tablets, and Chromebooks.
The web version is ideal for:
- Quick one-off text extractions from screenshots
- Digitizing scanned PDFs without desktop software
- Students and researchers working across multiple devices
Getting Started
Simply visit umiocr.com, click Launch OCR Tool in the hero section, upload your image, and extract text in seconds. No sign-up required.