OCR PDF
Read the text on scanned pages. Get the same PDF back with an invisible text layer, so you can search and copy from it, or save the words as a .txt file. The text recognition runs inside your browser, so your pages never leave your device.
- Free
- Unlimited
- No signup
- No watermark
- Files never leave your device
or drop it here
Text reading
Reading is done by your device, so it takes a few seconds per page. The language reader (a few MB) is downloaded from this site once.
File name
Protection
AES-256. If you forget the password, nobody can recover the file, including us.
How to make a scanned PDF searchable with OCR
- Drop your scanned PDF. If it is password-protected, type the password.
- Choose "Searchable PDF" or "Text file", and the pages to read. Pages that already have text are skipped.
- Click Make searchable PDF. Pick the language of the pages first. The first page takes longest because the language reader loads once (a few MB). Then download the result.
Questions
What does the searchable PDF look like?
Exactly like your original. PDFboat keeps every page picture untouched and adds the recognised words as invisible text on top, so Ctrl+F finds them and you can select and copy them.
Which languages work?
English, Spanish, French, German, Portuguese, Italian, Dutch, Russian, Hindi, Arabic, Chinese (Simplified), Japanese and Korean. Choose the language of your pages. For a document in two languages, keep "Also read English" on. Handwriting and mixed scripts in one line are read poorly.
How accurate is it?
Good on clean, straight, high-contrast scans, weaker on blurry photos, small print, columns of numbers and tables. It is a computer reading the page, so check important figures against the picture.
Why is it slow?
The reading is done by your own device, not a server. Expect a few seconds per page on a laptop. Keep the tab open while it works. "Standard (200 dpi)" is faster, "Detailed (300 dpi)" reads small print better.
My pages are sideways or upside down.
Turn them first with Rotate pages, then run OCR. Tilted photos work better after Scan to PDF straightens them.
Is my file uploaded?
No. The reader and its English data are files from this site, and the pages are read in your browser tab. Nothing is sent anywhere.
Can I password-protect the result?
Yes, for the searchable PDF. Switch on "Password-protect" and the PDF is encrypted with AES-256. It is off by default. A text file is not protected.