Extract text from images right in your browser.
Tesseract OCR, fully on-device via WebAssembly. Upload an image or paste a screenshot — 100+ languages, zero uploads.
Accurate text recognition. Zero uploads.
Tesseract.js running locally in your browser — no servers, no API keys, no limits.
100% private
Your images never leave your device. All recognition happens locally via WebAssembly.
100+ languages
English, Chinese, Japanese, Korean, and major European scripts — switch on the fly.
Word-level boxes
Every recognized word comes with a bounding box and confidence score, overlaid on the image.
Export anywhere
Copy plain text, or download as TXT, JSON (with positions), or searchable hOCR.
Three steps. Zero servers.
- 01
Upload or paste
Drag & drop an image, paste a screenshot, or browse. PNG, JPG, WebP, BMP, and GIF.
- 02
OCR runs locally
Tesseract recognizes text in your browser via WebAssembly, with word positions and confidence.
- 03
Copy or export
Copy the text or download as TXT, JSON, or hOCR. The engine caches for reuse.
About the OCR tool
A free OCR tool that extracts text from images and documents in your browser — screenshots, photos of whiteboards, scanned pages, certificates, and hand-printed notes. It is designed for anyone who would rather not upload sensitive documents to an online conversion service.
How it works
The tool runs Tesseract.js — the open-source Tesseract OCR engine compiled to WebAssembly — entirely client-side. Language data for 60+ languages downloads on demand and is cached after first use, including Chinese, English, Japanese, and Korean recognition packs. Images are preprocessed in-browser (grayscale, contrast, deskew hints), then recognized line by line. Results are selectable, copy-ready text you can export to a .txt file.
Limits & requirements
Recognition quality follows source quality: sharp, well-lit, straight images produce near-perfect text; blurry camera angles or stylized fonts need corrections. First use of a new language downloads its trained data. Heavy batches run fine but are bound by your device memory; WebGPU acceleration helps when available.
Privacy
Screenshots and scans of contracts, IDs, or private correspondence stay on your device. Nothing is uploaded to perform recognition — a meaningful distinction for documents you would not trust with a random conversion website.