How to OCR a PDF
Pick the way that fits your file: a browser tool for a quick private job, Google Docs if you want editable text, Acrobat if you already pay for it, OCRmyPDF for big batches.
Updated 12 November 2026
Key takeaways
- OCR turns pictures of text into real text you can search and copy.
- Some ways keep your PDF and add a text layer; others give you a new document with the text only.
- Online tools differ most in one thing: whether your file is uploaded.
1. In your browser, nothing uploaded
Use OCR PDF. Choose the PDF, pick the language, press Make searchable, and download the same PDF with a text layer, plus a .txt file if you want it. It runs on Tesseract.js, the browser version of the open-source Tesseract engine.
Best for: private papers, up to 50 pages, on any computer. Not for: whole books, or when you want to edit the text.
OCR your PDF here, free and private.
OCR my PDF2. Google Docs
In Google Drive, right-click the PDF and choose Open with, then Google Docs. Drive reads the text into a new document you can edit.
Best for: getting editable text out fast. Limits: Google asks for files of 2 MB or less, and says lists, tables, columns and footnotes are not likely to come through. You get a Doc, not a searchable PDF, and the file goes to Google.
3. Adobe Acrobat
In the paid desktop app, open the PDF, go to Tools, choose Recognize Text, and keep the output style Searchable Image so the pages look unchanged. Adobe also runs a free online OCR tool, which uploads your file.
Best for: people who already pay for Acrobat and need to edit the words afterwards.
4. On a Mac
Preview has no OCR that saves into the PDF. Live Text lets you drag across text in an image and copy it, which is handy for a sentence or two. For a searchable file, use method 1, 2 or 3; they all work on a Mac.
5. OCRmyPDF, for batches
OCRmyPDF is a free command-line tool that adds a text layer to PDFs using Tesseract. Once installed, one command does a file, and a short script does a folder.
Best for: hundreds of files, if you are comfortable in a terminal.
Which to pick
| Way | You get | File uploaded | Limit |
|---|---|---|---|
| OCR PDF | Same PDF, searchable, plus .txt | No | 50 pages a run |
| Google Docs | A new Google Doc | To Google | 2 MB file |
| Acrobat desktop | Same PDF, searchable or editable | No | Paid plan |
| Acrobat online | Searchable PDF | To Adobe | Not checked |
| OCRmyPDF | Same PDF, searchable | No | None stated |
| FixMyScan | Cleaned, searchable book PDF | Yes, deleted 24 hours after the book is ready | 1,000 pages |
Get better results from any of them
- Pick the right language. On a German page we tested, the English model got none of the 16 words with umlauts right (see the example).
- Scan at 300 dpi if you are making the scan yourself.
- Straighten tilted pages and crop dark edges first. Misreads cluster at crooked lines and borders.
Questions
What does OCR stand for?
Optical character recognition: software that looks at a picture of text and works out which letters are in it.
Can a Mac OCR a PDF without extra software?
Not into the file. Live Text lets you copy words from an image in Preview, but it does not save a text layer, so the PDF stays unsearchable everywhere else. Use this site or another OCR tool to save one.
Does Windows have built-in PDF OCR?
There is no built-in tool that adds a text layer to a PDF. Google Docs and the free tools on this page work on Windows.
Is free OCR accurate?
On a clean, straight, printed page, free engines such as Tesseract read most words correctly. Handwriting, faint print and tilted or curved pages are where every engine slips.
Can OCR read handwriting?
Not reliably with Tesseract, the engine this site uses. It is built for printed text.
How long does it take?
It depends on your computer and the page size. You see a progress bar per page, so you can tell before it finishes.