Turn text in images and PDFs into text, in your browser alone.

No sign-up, no installation, and your data is never sent to a server.
Vertical Japanese text is supported too. All character recognition happens entirely inside your own browser.

Free OCR (text recognition) tools that work entirely in your browser

Archipela OCR is a free collection of OCR tools that turn the text captured in images and PDFs into editable text. It runs the open-source OCR engine Tesseract, compiled to WebAssembly, inside your browser, so it can recognise text without uploading your images or PDFs to a server. No account and no installation required.

It handles Japanese (horizontal and vertical), English and mixed Japanese-English text, and includes tools built around the situations people actually run into: pasting a screenshot, transcribing a scanned PDF, and automatically pulling the invoice registration number, amount and date out of a receipt. Language data is loaded only on first use — that is a download to your device, and your images are never sent anywhere.

🛡️

Secure by design

Images and PDFs are never sent to a server; everything is processed inside your browser

📖

Vertical text supported

Vertically typeset Japanese documents can be transcribed as they are

💸

Completely free

Free to use, with no limit on runs or number of files

📱

Multi-device

Works on desktop, tablet and smartphone alike