Image to text (OCR)

Turn photos, screenshots and scanned PDFs into text you can copy and edit. The reading happens on your device.

Your files stay on your deviceFree, no sign-up

Before you use it

Works well for

  • Printed text in photos, screenshots and scans
  • Scanned PDFs (digital text is copied exactly)
  • Portuguese and English documents

Not the right tool for

  • Handwriting
  • Blurry, skewed or very dark photos — expect mistakes
  • Tables and columns: the text comes out, the layout doesn't

How to use it

  1. Drop one or more images, or a PDF.
  2. Pick the language of the text. Mixed documents can use both.
  3. Wait for the reading to finish, then fix anything that came out wrong and copy or download the text.

For PDFs, pages that already contain digital text are copied exactly, without OCR. Only scanned pages are read as images.

Getting the best results

  • Photograph the page straight on, in good light, without shadows.
  • Fill the frame with the text; avoid tiny text in a big photo.
  • Keep "Improve the image first" on — it sharpens contrast and enlarges small images.

The first time you use it, your browser downloads the reading engine and the language model (a few MB). After that it's cached and starts faster.

Frequently asked questions

Can it read handwriting?

Not reliably. It's built for printed and typed text. Very neat block letters sometimes work.

Is it 100% accurate?

No OCR is. Clean printed text is usually read almost perfectly; blurry, skewed or low-contrast photos produce mistakes, which is why the result is editable.

Are my files uploaded?

No. Both the engine and the language models run in your browser.