OCR PDF

PDF ToolsNewPremium

Read the text out of a scanned PDF, page by page.

Nothing Is UploadedRuns In This Browser TabNo Account Needed

How It Works

1

Upload / Input

Select your files or paste your content to get started.

2

Process

The work happens in this browser tab. Your file never leaves your device, so nothing is queued, stored or transmitted.

3

Download / Result

Get your results instantly. Download files or copy output in one click.

Features

Opens a scanned PDF directly and reads every page in order
Pages that already hold text are read exactly, not recognised
Also takes page images: JPG, PNG, WEBP, BMP and TIFF
Fourteen recognition languages, all served from this site
Per-page progress, and a long document can be stopped at any point
Plain-text result you can edit, copy or download as .txt
Runs in your browser — the document is never uploaded

How to use OCR PDF

  1. Drop the PDF in

    A scanned PDF goes straight in — there is no need to export it to images first. Individual page images work too, added in reading order.

  2. Pick the language

    Choose the language of the document. This matters: the wrong language setting produces markedly worse results.

  3. Watch it work, or stop it

    Progress is counted in pages, not guessed from a clock, and a long document can be stopped at any point without losing the page you are on.

  4. Read, edit and download the text

    The text appears in an editable box, labelled page by page. Fix anything the recognition got wrong, then copy it or download it as a .txt file.

When to use OCR PDF

  • Getting the figures out of a scanned invoice without retyping them
  • Copying a quotation out of a photographed page of a book or report
  • Pulling the text out of a scan so it can be pasted into a document
  • Turning a stack of scanned pages into one plain-text transcript

FAQ

Yes. Drop the PDF straight in and every page is read in order. A page with no text of its own is rendered at 300 DPI and recognised; a page that already carries a text layer is read from it directly, which is both exact and instant.
No. What comes back is plain text. A searchable PDF needs an invisible text layer written back into the original document, and nothing here writes one — the PDF you started with is untouched.
It is refused, with a message that says so, as soon as you drop it — not three minutes into a run. Remove the password with Unlock PDF first and then come back.
Fourteen: English, Spanish, French, German, Italian, Portuguese, Dutch, Russian, Arabic, Hebrew, Hindi, Chinese (Simplified), Japanese and Korean.
Accuracy depends on scan quality. Clear, high-resolution scans of printed text produce the best results. Pages read from an existing text layer are not recognised at all, so they carry no recognition error.
Yes. The file picker takes .pdf, the document is opened in your browser, and every page is read in order. It used to take images only, which was a strange thing for a tool of this name to do; that is fixed.
Because it already held text. A PDF page that carries a text layer is read from it rather than recognised, which takes no measurable time and is exact. The summary above the result says how many pages went each way.
It cannot be opened, and you are told exactly that when you drop it rather than after a long wait. Run it through Unlock PDF first if you have the password.
On a clean 300 DPI scan of printed text, typically 98% or better. Accuracy drops quickly with low resolution, skew, poor lighting, unusual fonts or background patterns. Always proofread anything that matters.
Not reliably. OCR is built for printed type. Handwriting recognition is a distinctly harder problem and results vary from poor to unusable depending on the handwriting.
The usual causes are resolution below 200 DPI, a skewed or curved page, poor or uneven lighting, a coloured or patterned background, or the wrong language selected. Re-scanning at 300 DPI in greyscale fixes most of them.

Related Tools

PopularNew

PDF to Image

Convert PDF pages to JPG, PNG, or WEBP images.

converters

OCR Online

Extract editable text from an image or a scanned PDF.

image-tools

Document Scanner

Turn a photo of a document into something that reads like a scan.

image-tools
Popular

Merge PDF

Combine PDF files into a single document in seconds. Merging several PDFs is Premium.

pdf-tools

About OCR PDF

OCR — optical character recognition — turns a picture of text into actual text. A scan looks readable to you but is opaque to a computer: searching it finds nothing and selecting text is impossible. Drop the PDF in and every page is rendered at 300 DPI and recognised in turn, in your browser, with nothing uploaded anywhere.

Not every page of a PDF needs that. Plenty of documents are already text — exported from a word processor, or scanned by something that ran its own OCR and saved the result behind the image. Those pages are read from the text layer instead of recognised, which is both instant and exact rather than approximate, and the tool says on screen how many pages went each way.

What it does not do is write that text back into a PDF. A genuinely searchable PDF carries an invisible text layer positioned behind the page image, and producing one means re-writing the document. Your PDF is read and never modified. If a searchable PDF is what you need, this tool gets you the text and no further.

Accuracy depends almost entirely on input quality. A clean 300 DPI scan of printed text typically comes back well above 98% accurate. A 150 DPI scan, a skewed page, or a phone photo taken in poor light and accuracy falls sharply. Handwriting is a different problem entirely and should not be expected to work.

Tips and limits

  • Scan at 300 DPI. Below 200 DPI accuracy degrades badly and no software fixes that.
  • Set the correct language. Running English OCR over a French document produces nonsense.
  • Straighten skewed pages first — even a few degrees of rotation costs accuracy.
  • Handwriting is not reliably recognised. OCR is designed for printed text.
  • Greyscale or black-and-white scans usually OCR better than colour ones.
  • A long document is minutes of work on your own machine. Start it, and stop it if you change your mind — nothing is queued anywhere else.