How to use OCR PDF
- Choose the scanned PDF.
- Tick the languages it's written in; English is selected to start.
- Select Make searchable. Each page is read in turn.
- Download the searchable PDF, and copy or save the recognised text if you need it.
What OCR PDF does
A scanned PDF is a stack of photographs of paper. It looks like a document, but to a computer each page is just pixels: you can't search it, select a sentence, copy a figure or have it read aloud. OCR, optical character recognition, reads the letters in those pictures and turns them back into text.
This tool renders each page, reads it with Tesseract, a widely used open-source OCR engine, and lays the recognised text invisibly over the page, exactly where the words appear in the scan. The result looks identical to your original, because the scanned images aren't touched or re-compressed, but you can now search it with Ctrl+F, select and copy text, and find it later in your file search.
Private by design
Scanned documents are often the sensitive ones: signed contracts, ID cards, medical letters, bank forms. Online OCR services ask you to upload them. Here the whole process runs in your browser. The OCR engine and the language data, about 4 to 7 MB in total, download once from this site and your browser keeps them, so later runs start at once. The PDF itself never leaves your device.
Getting good results
Tick only the languages the document uses: each extra language slows recognition and adds chances to misread. Clean, straight scans of printed text at a normal size read very accurately. Phone photos taken at an angle, faint photocopies, handwriting and decorative fonts read less well, so check names, dates and amounts before relying on them.
Pages that already have a text layer, such as a typed cover letter in front of scanned attachments, are left as they are by default, which is faster and keeps their exact text. Turn the option off to read every page again.
Speed
On a laptop a typical page takes one to three seconds; phones are slower. Progress is shown page by page, and you can keep using other tabs while it works.
Questions
- Does the PDF look different afterwards?
- No. The scanned pages are kept exactly as they were, without re-compressing them. The recognised text is added as an invisible layer on top, which is what makes it searchable and selectable.
- Is my PDF uploaded for OCR?
- No. Text recognition runs in your browser with Tesseract, an open-source OCR engine. The engine and language data download once from this site and are kept by your browser; the PDF never leaves your device.
- How accurate is it?
- Clean, straight scans of printed text read very accurately. Handwriting, low-resolution photos, skewed pages and decorative fonts read poorly, so check names and figures before relying on them.