PDF OCR
Read the text in scanned PDFs into a .txt file, page by page, on your own device. English. No upload, no signup, no watermark.
Files stay on your device How it works
Drop files here, paste, or
PDF · up to 5 files, 100 MB each
Your files
Your file is ready
The words inside a scan
A scanned PDF is a stack of pictures of pages. It looks like a document, but you can’t select a line, search for a word or paste a paragraph into an email. This tool reads each page with OCR (optical character recognition) and gives you the text as a plain file, so an old contract, a scanned letter or a photocopied chapter becomes something you can search and quote.
How to use it
- Choose up to 5 PDFs, each up to 100 MB and up to 50 pages.
- Keep Mark where each page starts ticked to see which page each passage came from.
- Press Read the text. A progress bar shows the page being read; Cancel stops at once.
- Download each .txt file, or all of them as a ZIP.
How it works
Your browser draws each page at 300 dots per inch, the usual resolution for scanning documents, and passes the
picture to Tesseract, the open-source OCR engine, running in a separate worker on your device. The words it finds
go into the text file in reading order, with a --- Page 3 --- line before each page when marking is on.
A page where nothing could be read keeps its marker with a note, so the numbering still matches the PDF.
For the best result
- Straight, clean scans read best. If pages are sideways, turn them first with Rotate PDF.
- Long scans: split them into parts of up to 50 pages with Split PDF.
- Only need a few pages? Take them out first with Extract PDF Pages; it’s quicker.
Good to know
- If you can already select the text in your PDF viewer, the PDF has real text: PDF to Text is faster and exact.
- For a photo or screenshot, use Image to Text.
- English only for now. Accented letters may lose their accents.
- Password-protected PDFs need the password removed first.