Make a Scanned PDF Searchable

Run text recognition (OCR) on a scanned PDF so you can search it, select text and copy it. The pages look the same as before.

Runs in your browser. Your files never leave your device.

Make a Scanned PDF SearchableRuns on your device
Scanned PDF

Page numbers start at 1.

Your PDF is processed on your device and is never uploaded. The text recognition engine and its language data load from this site the first time you start, which can take a minute on a slow connection.

At a glance

Runs
In your browser, on your device
Works with
PDF
Largest file
50 MB
Cost
Free, no account needed

What Make a Scanned PDF Searchable does

Make a Scanned PDF Searchable runs text recognition (OCR) on a scanned PDF. Each page is read, and the words are added as an invisible text layer behind the picture, so the page looks exactly the same but can be searched, selected and copied.

Recognition runs in your browser with the open-source Tesseract engine. The engine and its English language data load from this site the first time you start, then your browser keeps them for next time.

How to use Make a Scanned PDF Searchable

  1. Choose the scanned PDF.
  2. Pick the language of the text.
  3. Select Recognise text. Each page takes a few seconds on a computer and longer on a phone.
  4. Download the searchable PDF and try searching it for a word you can see.

Examples

An archive of old letters

Scanned letters are pictures, so a search for a name finds nothing. After recognition, searching the PDF finds every page that mentions it.

Privacy and limits

The PDF is opened and rebuilt inside your browser tab. It is not uploaded to our server or to anyone else, so it is safe to use with contracts, statements and other private documents. Closing the tab clears it from memory. The text recognition engine runs inside your browser too; no page is sent anywhere to be read.

  • English is supported for now.
  • Handwriting, very small print and low-resolution scans are read poorly.
  • Up to 100 pages at a time. Pages that already contain text are skipped.
  • The first run downloads about 6 MB (the engine and English data).

Questions

Will the pages look different?
No. The original page pictures are kept; only an invisible layer of text is added behind them.
How accurate is it?
For clean, printed text it gets the large majority of words right. Names, numbers and anything faint or skewed deserve a check before you rely on them.
Why was a page skipped?
It already had text, so reading it again would only add a second copy.
Can I get the text on its own?
Yes. Run PDF to Text on the searchable PDF to copy or save the recognised text.
Which languages can it read?
English for now. Other languages need larger data files and fonts that can show their scripts in the PDF.

Further reading

All PDF tools