Searchable PDF
Make a scanned PDF searchable: text recognition adds a selectable text layer to a new PDF, processed on your device. Files stay in your browser.
Recognition files ≈ 26 MB on first use, plus language data.
Recognized text
Was this tool useful?
Try an example
Download sampleSelect English. Download the PDF and search for INVOICE, 148.50, DELIVERY and 2048. Both pages should look like the sample, with text you can select and copy.
What to expect
Each page is recognized as an image, in order. English, Russian and mixed recognition are available. OCR can misread letters, numbers, columns and tables: review the result. This creates a new PDF from rendered page images and an invisible text layer. Visible page size and rotation are retained; forms, links, bookmarks, attachments and digital signatures are not retained. Existing text is recognized again; image quality is limited by rendering resolution.
Pages are rendered up to 4 megapixels each. Loading and each processing step allow up to 90 seconds; the whole job stops after 10 minutes. Blank pages are marked in the text. No download is offered if processing fails or no text is recognized.
Protected PDF?Unlock it with its password first.
The first run may download converter files and take a moment. If it fails, check your connection or try a smaller file. Cancel stops active processing; you can choose another file and retry.
Engine licenses and source code
A practical guide
Add searchable text to a scan.
- Open a scanned PDF and select the language of its text.
- Run recognition and wait until all pages are processed.
- Download the new PDF and search it for a word you can see on the page.
Try this example
Process a two-page scanned invoice in English, then search the result for the total and confirm you can select and copy it.
Before you start: Recognition depends on scan quality and the chosen language, so handwriting and faint print can be misread. Compare numbers with the original.