Free OCR
OCR extracts editable text from photos, screenshots and scanned PDFs with the open-source Tesseract engine, in 19 languages that can be combined. Improve faint or tilted scans, fix sideways pages, review words marked as uncertain, then download a searchable PDF, a Word file, plain text or JSON. Several files are read as one document.
- Languages:
- 19, combinable
- Output:
- Searchable PDF, Word, TXT, JSON
- Engine:
- Tesseract.js 7
- Price:
- Free
- Runs:
- Locally, in your browser
- Uploads:
- None: files never leave your device
- Account:
- Not needed
Quick answers
Is my document uploaded?
No. Tesseract, the open-source OCR engine, runs inside your browser as WebAssembly. Only the engine and the language models are downloaded from this site; your files never leave your device.
What is a searchable PDF?
A PDF that looks exactly like your scan but has an invisible layer of recognized text on top, so you can search it, select and copy words, and have it read aloud. Pages of an existing PDF are kept as they are.
Processed on your device: your files are never uploaded. How it works
Network activity in this tab
What this page has requested since you opened the site in this tab. An upload is any request that sends data: POST, PUT, PATCH or a request with a body.
No uploads: nothing has been sent from this page.
No network requests yet.
To see every request, open your browser's developer tools (Network tab).
How to use the OCR
- Drop photos, screenshots or scanned PDFs (several files are read as one document, in order).
- Choose the languages of the text and, if needed, the cleanup options, then click Recognize text.
- Correct the text if you like, check the highlighted low-confidence words in the page preview, and download a searchable PDF, Word file, plain text or JSON.
Credits & open source
Built on open-source libraries. Thank you to the people who make these projects.
- Tesseract.jsTesseract OCR engine compiled to WebAssembly. © Project Naptha (Jerome Wu, Kevin Kwok, Guillermo Webster); Tesseract OCR by Google and contributors · Apache-2.0
- PDF.jsRenders PDF pages to images and thumbnails. © Mozilla Foundation and contributors · Apache-2.0
- pdf-libCreate and modify PDF documents (merge, split, rotate, stamp, sign). © Andrew Dillon · MIT
- docxCreates Word .docx files. © Dolan Miu · MIT
Web platform APIs used: File API, Canvas 2D, createImageBitmap, Web Workers, Clipboard API, URL API.
- tessdata (best_int LSTM models) by Google / Tesseract OCR contributors (Apache-2.0) — OCR language models
Full license texts: third-party notices · all credits
Frequently asked questions
How can I get better results?
Select the right languages (and only those), use scans of at least 300 DPI, and try Black & white for faint or colored backgrounds. Turn on “Fix sideways or upside-down pages” for rotated scans. Words highlighted in the preview are the ones the engine was unsure about.
Does it read handwriting?
Only neat, print-like handwriting, and not reliably. Tesseract is trained on printed text.
Which languages are supported?
English, Spanish, French, German, Italian, Portuguese, Dutch, Polish, Turkish, Indonesian, Vietnamese, Russian, Ukrainian, Arabic, Hindi, Chinese (Simplified and Traditional), Japanese and Korean. You can combine several in one document.