OCR a scanned PDF
Make a scan searchable
Where it runs: On the server.
Settings
Drop a file to begin.
This tool needs software that does not run in a browser. The file is sent over an encrypted connection, processed, then deleted automatically after two hours. Nobody reads it.
What this tool does
An invisible text layer is placed exactly under the image: the document keeps its appearance but becomes selectable, searchable and readable by a screen reader. Eight languages recognised.
In the same family
Where the file goes
How to do it
- 01Drop in the scanned PDF.
- 02Choose the language of the document.
- 03Run the OCR and download a searchable document.
Frequently asked questions
- Does the document change appearance?
- No. An invisible text layer is placed exactly under the image: the document keeps the precise look of the scan but becomes selectable, searchable and readable by a screen reader.
- Which languages are recognised?
- Eight, chosen before the run. The choice matters more than you might think: recognition leans on a dictionary, and French text treated as English comes back riddled with errors on the accents.
- Does a poor scan give a good result?
- Rarely. A page that is skewed, stained or scanned below 200 dots per inch limits what recognition can do. Tick “Straighten skewed pages”, trim the margins with Crop, and rescan at a higher resolution if you can.