Extract text from PDFs (OCR)

Read the text on each page of a PDF, then copy it or save it as a .txt file. Nothing is uploaded to an outside server.

or drop it here

Files are processed on your device and never sent to an outside server.

The first time only, about 38 MB of recognition models is downloaded. On devices without WebGPU, a page can take 10 seconds or more. Keep this tab in front while it works.

How to use

  1. Choose the language of the text: English, Japanese, Chinese or Korean. To read only some pages, type them, such as 1-3, 5.
  2. Click “Choose file” or drop your PDF onto the box. Each page is drawn as an image and read right away.
  3. Check the text of each page, then copy it or save it as a .txt file. You can also save the text of all pages in one file.

Frequently asked questions

How accurate is it?

Printed text in clear scans is usually read almost exactly. Handwriting, decorative or very small text, faint scans, skewed photos and complex tables are harder, so check the result before you use it. You can correct the text in the box before copying or saving it. If you can already select the text, as in a PDF exported from Word, copying it in your PDF app is exact.

Can it read vertical Japanese text?

Yes. Vertical lines are read top to bottom, starting from the right. Punctuation that turns in vertical writing, such as 、 。 「 」, may be missing.

Why does the first page take longer?

The first time, your browser downloads the text recognition models (30 to 40 MB). They stay in the browser’s cache, so later pages start right away, even offline. On devices without WebGPU, the models run on the processor and a page can take 10 seconds or more; for a long PDF, type only the pages you need. Keep this tab in front while it works.

Which languages can it read?

English, Japanese and Chinese (Simplified and Traditional) are read by the same model, so whichever of them you choose, text that mixes them is read as it is. Korean is read by its own model: choose Korean for Korean text, and one of the others for everything else. Other languages written in the Latin alphabet may be read only in part.

Are my files uploaded to an outside server?

No. Everything runs in your browser, so your files never leave your device. The page only downloads the tool's own code when you add the first file.

Related tools