OCR PDF – extract text from scanned PDFs, free

4 min read

You want to copy the text from a PDF, but selecting doesn't work – because it's a scan or an image PDF? This guide shows you how to extract text from any PDF for free, right in your browser, including OCR for scanned pages – no software and without your file being uploaded.

Digital text vs. scan – why OCR?

A PDF can hold text in two ways: as digital text (selectable, copyable) or as an image (a scan or photo where the text is only "pictured"). Digital text can be read directly. Scans need OCR – optical character recognition, which turns the pixels back into real text.

Step by step: extract text from a PDF

Step 1: open the tool

Open the PDF-to-text tool. No login, no installation.

Step 2: add the PDF

Drag your PDF into the area. The tool first tries to read the digital text automatically – that's instant and fully local.

Step 3: run OCR for scans

If the tool finds little or no digital text (typical for scans), it offers OCR. Pick the document's language and start recognition. For a single snippet you can also just select a region and recognise only that.

Step 4: copy or download

Copy the recognised text to the clipboard or download it as TXT, RTF or DOCX.

Supported languages

OCR supports many languages (including English, German, Spanish, French, Danish, Portuguese, Polish). Choosing the language that actually appears in the document markedly improves recognition accuracy.

Common use cases

• Make text from scanned documents or contracts copyable.
• Pull content out of photo PDFs or screenshots.
• Turn old, non-searchable PDFs into editable text.

Privacy note: your PDF never leaves the browser. For recognition a small open-source language model (~4 MB) is downloaded to your device once – your file itself is never transmitted. No upload, no server.

Frequently asked questions

What is OCR?

OCR turns text inside images or scans into real, copyable text – needed for PDFs with no digital text.

Is it free?

Yes – completely free and local in the browser, with no registration, no subscription and no upload.

Does it work on scans and photos?

Yes. With little digital text, the tool offers OCR; you choose the language and the text is recognised from the page images.

Does my file stay private?

Yes. The file is not uploaded; only a small, open-source language model is downloaded to your device.