How to OCR a scanned PDF
- Drop your scanned PDF into the box above.
- Choose render quality — higher is more accurate but slower.
- Optionally limit pages (default: the first 5, since OCR takes a moment per page).
- Click Recognize text, then copy the result or download it as a .txt file.
Good to know
- OCR adds a text copy — it does not convert your scan into an editable Word document, and it does not modify the original PDF.
- Already have a text layer? Use the Extract Text tool instead — it is instant and pixel-perfect.
- Clean, high-contrast scans (300 dpi, straight pages) recognize best. Photos of screens or skewed pages reduce accuracy.
- For best results on a long document, process it in batches of 10–20 pages.
Frequently asked questions
Why is my PDF already selectable?
Many "scanned" PDFs actually contain a hidden text layer from the original scanning software. Try selecting text with your cursor — if it works, use Extract Text and skip OCR entirely.
Can I get a searchable PDF instead of a text file?
Not with this tool — it produces a plain text copy. Creating an invisible text layer inside a PDF is a heavier operation that is on the roadmap.
Does it work on my phone?
Yes, though recognition is slower on mobile processors. Stick to small page batches.