A scanned PDF is really a set of pictures of pages, so you can't search it or copy a paragraph out of it. OCR PDF reads the text in those pictures and adds it to the file as a hidden layer. You get a searchable PDF that looks the same as before, and you don't need to sign up or pay.
How to make a PDF searchable
- Upload your scanned PDF. Drag it in on a computer, or tap "Select files" on a phone. The tool takes one PDF of up to 50 pages.
- Choose the language the document is written in.
- Start the OCR. Give it a little time, since each page takes a few seconds.
- Download the searchable PDF and try searching for a word you can see on a page.
What OCR adds to your file
OCR stands for optical character recognition. The tool looks at each page image, recognises the letters and words, and places that text in an invisible layer inside the PDF.
The page images stay exactly as they were, so stamps, signatures, logos and handwriting all look the same as in your scan. What changes is what you can do with the file. In your PDF reader you can now press Ctrl+F (or use search on your phone) to find a word, select lines of text, and copy them into an email or a Word document.
Languages
Pick the language the document is written in, so the tool knows which letters and words to look for. The list includes:
- English, French, Spanish, Portuguese, German, Italian and Dutch
- Arabic, Hindi, Russian and Turkish
- Chinese (Simplified) and Indonesian
A few more are available in the menu on the tool. Pick the main language of the document.
What affects accuracy
OCR is a best attempt at reading the page, and its accuracy depends on the scan. Expect good results from clear, straight, well-lit scans of printed text. Expect more mistakes from:
- blurry or dark phone photos
- pages scanned at an angle or with shadows across them
- very small print or faded photocopies
- handwriting, which OCR reads poorly
If you're scanning a document yourself, take a little care at the start: flat pages and good light, with the phone held straight above the paper. OCR can only read what the scan shows clearly.
Pages scanned sideways should be turned upright first. Rotate PDF turns every page of a file at once.
Who uses OCR
Students and researchers
Old journal articles, library scans and photocopied handouts are often image-only PDFs. Making them searchable lets you jump straight to a term instead of reading 40 pages to find one quote, and lets you copy passages into your notes.
Offices and small businesses
Scanned contracts, invoices and receipts are much easier to work with when you can search them for a name or an invoice number. Run each file through OCR before you store it, and finding things later gets simpler.
Getting text out of a scan
If what you really want is the text itself, OCR is the first step. A scanned PDF has no text for a converter to pull out, so make the PDF searchable here, then use PDF to text for a plain .txt file or PDF to Word to edit it as a document.
Limits
- One PDF at a time, up to 50 pages and 100 MB. For a longer document, split it into parts with Split PDF and run each part through OCR.
- It's slower than the other tools on Xon File, at a few seconds per page. A 50-page file needs some patience.
- Accuracy depends on scan quality, and handwriting gives poor results.
- The page images are left as they are. The tool adds the hidden text and doesn't straighten or clean up the scan.
Privacy
Your PDF is processed on our server and deleted as soon as the searchable version is sent back to you. Nothing is stored or shared. No watermark or branding is added to your pages.
Frequently asked questions
Is OCR PDF free?
Yes. You can use it free of charge with no sign-up or account, and the result has no watermark.
How do I know if my PDF needs OCR?
Open it and try to select a word or search for one. If you can't select any text and search finds nothing, the pages are images, and OCR will help.
Will OCR change how my PDF looks?
No. The page images stay the same. The text is added as a hidden layer underneath, so the file looks identical but can be searched and copied.
Can it read handwriting?
Not well. OCR works best on clear printed text. Handwritten notes usually come out with many errors, so don't rely on the hidden text for handwritten pages.
Why is OCR taking longer than other tools?
Reading text from images is slow work, and it takes a few seconds for each page. A document close to the 50-page limit will take noticeably longer than a short one.
Can I OCR a PDF on my phone?
Yes. The tool runs in the browser on Android phones and iPhones, so there's nothing to install. Tap "Select files", choose your scanned PDF and its language, and keep the page open while it works through the pages. The searchable PDF then downloads to your phone.
Can I use OCR on a photo instead of a PDF?
This tool works on PDFs. For a single photo or screenshot, use Image to text on Xon File, which reads the text and gives you a .txt file.