OCR a PDF — make scans searchable
Turn scanned pages into searchable, copyable text, privately.
or drop it here · printed text in 12 languages
Processed in your browser. Your files are never uploaded. How we know
How to make a scanned PDF searchable
A scanned PDF is just a stack of pictures: you can’t search it, copy a sentence from it, or let a screen reader read it aloud. OCR (optical character recognition) reads the printed words in those pictures. This tool gives you either a searchable PDF that looks exactly like the original but has hidden, selectable text, or the recognised words as a plain text file. It supports 12 languages, up to three in one document.
- 1
Open your scan
Choose a scanned PDF. Pages that already have text are marked.
- 2
Pick the language
Choose up to three languages used in the document, and the output type.
- 3
Download
Get a searchable PDF or a .txt file, or copy the text.
OCR PDF features
- Searchable PDF or plain-text output
- 12 languages, up to three in one document
- Pages that already contain text are skipped
- Force OCR option for every page
- Average confidence shown when recognition finishes
- Recognition runs on your device, in background workers
OCR without uploading: what happens to your scan
Recognition runs on your own device using Tesseract, an open-source OCR engine, in background workers. Your document is not uploaded.
The first time you use a language, its recognition data (roughly 0.6 to 3 MB) is downloaded from our own site and kept in your browser’s cache. That download contains no information about your file. Each page is drawn at 300 dpi (less for very large pages, to fit your device’s memory), read, and then an invisible text layer is laid exactly over the words. The page image itself is not changed.
Pages that already contain real text are skipped, so born-digital PDFs aren’t processed twice. Tick Force OCR on all pages to override.
Making a scanned contract searchable
A 6-page contract was scanned to PDF in English. Open it: all six pages are marked “Will be read”. Keep English selected, choose Searchable PDF and press OCR 6 pages. A progress bar shows which pages are being read and estimates the time left; on a laptop it usually takes a few seconds per page, longer on a phone. When it finishes, you’ll see the average confidence. Open the result, press Ctrl+F, search for the landlord’s name, and it’s found on page 1.
Good to know
- OCR works on printed text; handwriting isn’t supported. Accuracy depends on the scan, so check names, numbers and dates against the original.
- Scans at 300 dpi, straight and evenly lit, give the best results. If the result warns that the scan looks blurry or small, try rescanning.
- Pick every language in the document, but no more: each extra language makes OCR a little slower.
- Making a PDF from photos first? Use JPG to PDF, then OCR it, then compress it if it’s large.
Related tools and guides
OCR PDF: frequently asked questions
Does it read handwriting?
No. OCR here works on printed text. Accuracy depends on scan quality, so check anything important.
Which languages are supported?
English, Spanish, French, German, Portuguese, Italian, Dutch, Polish, Turkish, Indonesian, Hindi and Simplified Chinese. You can pick up to three at once.
Is my document uploaded for OCR?
No. Text recognition runs on your device. Only the language data is downloaded from our own site, and that request contains nothing about your file.
Is the original image kept?
Yes. A searchable PDF keeps each page image exactly as it was and adds invisible text over it, so it looks identical.
How long does it take?
It depends on your device and the number of pages. A few seconds per page on a modern laptop is typical; phones are slower. You can cancel at any time.