How to Get Text from a Scanned PDF (Free OCR)

If you can't select or copy anything in a PDF, it's a scan: each page is a photo of paper, with no text inside. To get the words out, use OCR (optical character recognition). Open PDF OCR, add the scanned PDF, choose the document's language, and click Start OCR. Every page is read in turn, and you can copy the text or download it as a .txt file.

How to OCR a scanned PDF

1. Open PDF OCR and add your PDF. You can also add photos or screenshots of documents.
2. Choose the language the document is written in — this matters more than any other setting.
3. Click Start OCR. Each page is drawn at about 200 dpi and read; the progress is shown page by page, and you can stop at any time.
4. Copy the text, or download it as a .txt file. Multi-page results are labelled "Page 1", "Page 2" and so on.

The first run downloads the recognition engine and the language data, so it takes a little longer. After that it starts straight away.

Is my PDF a scan?

Three quick checks:

• Try to select a word. In a scan, you select the whole page like an image, or nothing at all.
• Search for a word with Ctrl+F (Cmd+F on a Mac). A scan finds nothing.
• Zoom right in. Scanned text goes soft and pixelated; real text stays sharp at any zoom.

If the PDF does contain real text, PDF to Text is faster and exact. You don't have to decide in advance, though: PDF OCR detects pages that already contain text and reads those directly, and only runs OCR on the scanned ones.

Getting the most accurate results

OCR quality depends mostly on the scan:

• Choose the right language. The engine uses it to recognise letters and common words.
• Straight, flat pages read far better than curled book pages photographed at an angle.
• Even lighting matters: shadows across a page, or a phone's shadow, cause missed words.
• Sideways pages should be turned upright first with Rotate PDF.
• Clean, printed text is ideal. Handwriting is read poorly by any free OCR; neat block capitals work best.
• Small print needs a sharper scan. If you can rescan, 300 dpi is the sweet spot for documents.

Always proofread the result. OCR confuses similar shapes — 0 and O, 1 and l, rn and m — and numbers in invoices or IDs are worth double-checking.

What you can do with the text

• Paste it into Word or Google Docs to edit and reuse a letter you only have on paper.
• Search a long scanned contract for a clause.
• Translate it: paste the text into the AI Translator.
• Quote a passage from a scanned book or article without retyping.

A single screenshot or photo?

For one image — a screenshot of an error message, a photo of a sign — Image to Text is a lighter tool with clean-up controls, and works well for clear, typed text on a plain background.

Frequently asked questions

Is my document uploaded?

No. Recognition runs in your browser on your own device, which makes it suitable for contracts, ID documents and medical records.

Does OCR make my PDF searchable?

This tool gives you the text to copy or save as a .txt file; it doesn't rewrite the PDF itself. Keep the text file alongside the PDF, or paste it into a document.

How long does it take?

A few seconds per page on a modern computer; longer on a phone or for very dense pages. You can stop at any point and keep the pages read so far.

Which languages are supported?

29, including English, Spanish, French, German, Chinese, Japanese, Korean, Arabic, Hindi, Thai and Vietnamese. See how to OCR documents in other languages for tips.