How to make a scanned PDF searchable (free OCR in your browser)
Updated · PDF Converter
A scanned contract, a photographed receipt or an old fax saved as PDF looks like a normal document, but to your computer each page is just a picture. Search finds nothing, you cannot select a sentence, and copying text gives you an empty clipboard. OCR (optical character recognition) fixes this by reading the letters in the image and adding real text to the file. This guide shows how to do it for free, without installing anything.
The quick way: run OCR on the PDF
- Open OCR PDF and add your scanned PDF. You can add several files at once.
- Under Document language, pick the language the document is written in – for example English, Spanish or German. Mixed documents can use English + Spanish or English + French.
- Leave Recognition resolution on Balanced (200 dpi) unless you have a reason to change it (see below).
- Keep Skip pages that already have text ticked.
- Press OCR PDF and download the searchable PDF.
Open the result and press Ctrl+F (Cmd+F on a Mac): words on the scanned pages can now be found, selected and copied.
What OCR changes – and what it does not
The tool runs the Tesseract recognition engine on each page and then places an invisible text layer exactly over the words in the image. The page images themselves are not changed, so the document looks the same as before. You just gain the ability to search, select and copy.
The option Skip pages that already have text is useful for mixed documents – for example a report where most pages were exported from Word but a few signed pages were scanned in. Pages that already contain real text are left alone, which saves time and avoids adding a second text layer on top of the first.
Choosing the language and resolution
Picking the right language matters more than anything else. Recognition models know the letters, accents and common words of their language, so a French document read as English will come out with missing accents and more mistakes. More than 20 languages are available, including Arabic, Hindi, Bengali, Chinese (Simplified and Traditional), Japanese, Korean, Russian and Ukrainian.
| Recognition resolution | When to use it |
|---|---|
| Fast (150 dpi) | Long documents with large, clear print where speed matters |
| Balanced (200 dpi) | The default – a good choice for most scans |
| Accurate (300 dpi, slower) | Small print, footnotes, dense tables or slightly faint scans |
Tip: if words come out wrong, try again with Accurate (300 dpi). It takes longer, but small letters are recognised more reliably.
Getting better OCR results
- Start with a clean scan. Accuracy is best on straight, sharp scans at 200–300 dpi. Crooked pages, shadows and blur all cause errors.
- Printed text works best. Handwriting is not reliably recognised.
- Photos of documents? If you only have phone pictures, combine them into one PDF first with JPG to PDF, then run OCR on that PDF.
- Need the text only? For a single screenshot or photo, Image to Text recognises the words and lets you copy them or download a TXT file, without making a PDF.
- Need to edit the document? A searchable PDF is still a PDF. To rewrite the content, use PDF to Word with OCR turned on to get an editable Word file.
Is it private?
Yes. Recognition runs inside your browser, so the document itself never leaves your device. English recognition data is built in. For other languages, the recognition model (2–15 MB) is downloaded once from the jsDelivr CDN – that download contains only the language data, not your file.
Speed depends on your device: expect a few seconds per page on a modern laptop. Long scans take a while, so keep the tab open until the download is ready. If the finished file is large, you can shrink it afterwards with Compress PDF.
Frequently asked questions
How do I know if my PDF is scanned or already searchable?
Open it and try to select a word with the mouse, or search for a word you can see on the page. If nothing gets highlighted, the page is an image and needs OCR.
Will OCR change how my PDF looks?
No. The original page images stay exactly as they were. The recognised text is added as an invisible layer on top, so only search, selection and copying change.
Can I OCR a PDF on my phone?
Yes. Open the OCR page in a mobile browser, choose the PDF from your files and run it. Recognition happens on the phone itself, so it is slower than on a laptop, especially for long documents.