OCR PDF

Make scanned PDFs searchable with text recognition

pdfty's OCR tool turns image-based PDFs (scans, photos of documents) into searchable, copyable text. We use Tesseract 5 with optimized models for English, Russian, French, German, and Spanish — including Cyrillic and accented characters.

After OCR your PDF looks the same visually but you can now copy text, search, and convert to Word with the original layout preserved. Useful for scanned contracts, archived documents, photographed whiteboards, or fax outputs.

Free for files up to 20 MB, with no daily limit. Pages with mixed scripts work — OCR detects script per region. We never train models on your data.

How it works

  1. 1

    Upload scan

    Drop a scanned or image-only PDF. Up to 20 MB on free.

  2. 2

    Pick languages

    Choose one or more from English, Russian, French, German, Spanish.

  3. 3

    Run OCR

    Tesseract processes each page. Takes 1-3 seconds per page typically.

  4. 4

    Download searchable PDF

    Same look, with a hidden text layer. Now you can copy, search, convert.

  5. 5

    Done

    Files deleted within 1 hour.

Frequently asked questions

On clean 300 DPI scans, 98%+ for Latin scripts and 95%+ for Cyrillic. Lower DPI or noisy scans drop accuracy.
English, Russian, French, German and Spanish.
Tesseract handles printed text only. Handwriting recognition is on our roadmap.
Yes — pick all relevant languages in the picker. Tesseract auto-detects per region.
No. The original page image stays. We add an invisible text layer behind it.
Yes. OCR is free for any scan up to 20 MB; long scans just take a little longer.

Related tools

Guides & tutorials