Skip to content
Convert tools

PDF OCR — Extract Text from Scanned PDF, Free Online

Free PDF OCR: extract text from scanned or image-only PDFs, up to 20 pages per run, right in your browser. The OCR engine loads from a CDN on first use.

Convert in your browser · No sign-up

PDF up to 30 MB. OCR runs on up to 20 pages at a time.

Images (PNG, JPG, WebP, AVIF) work too: drop one here.

You can also paste a screenshot with Ctrl+V (⌘V on Mac).

On first use your browser downloads the OCR engine and the selected language data (several megabytes) from a public CDN and caches them. Your image itself is never uploaded.

How to use

  1. Choose a scanned PDF (up to 30 MB).
  2. Select the text language and the pages to read, for example 1-3, 5 (up to 20 pages per run).
  3. Click Extract text, then edit, copy or download the result as a .txt file.

Example

A 3-page scanned contract, pages 1-3 → "--- Page 1 ---\nAGREEMENT\nThis agreement is made…\n\n--- Page 2 ---\n…" as editable text, saved as contract.txt.

Details and limitations

Pages are rendered at about 200 DPI and recognized one after another with tesseract.js in your browser; the PDF is not uploaded. The OCR engine and language data are downloaded from a public CDN (jsDelivr) the first time and cached.

If the PDF already contains selectable text, copy it directly instead; OCR is for scanned or image-only pages. Password-protected PDFs are not supported, and output is plain text without layout.

FAQ

How do I extract text from a scanned PDF for free?

Choose the PDF, pick the language and the pages (up to 20 per run) and click Extract text. It is free, needs no account and runs in your browser. For longer documents, run it again with the next page range.

Can it read Japanese or Chinese PDFs with vertical text?

Yes. Choose Japanese (vertical), Simplified Chinese (vertical) or Traditional Chinese (vertical). For horizontal text use the normal Japanese or Chinese option.

Why is the result empty or full of mistakes?

OCR needs clear, printed text. Low-resolution scans, handwriting, unusual fonts or the wrong language setting reduce accuracy. Check that the language matches the document and try a cleaner scan.

Format references

Tool page updated: