OCR: extract text fromPDF & images

Extract text from a scanned PDF, a photo or a screenshot.

Drop your PDF or image here…
  • Private
  • Free
  • Tesseract.js

Need to copy text from a scanned PDF? This tool uses Tesseract OCR in your browser to read each page and extract the text. The file never leaves your device. Works with scanned PDFs, photos of documents and screenshots.

How does it work?

  1. Upload your file

    Drop a scanned PDF or an image (JPG, PNG, WebP). It opens in your browser — nothing is uploaded.

  2. Choose the language

    Select Spanish, English or both for better recognition accuracy.

  3. OCR with Tesseract

    Each page is analyzed with Tesseract.js, Google's open-source OCR engine.

  4. Copy or download the text

    The recognized text is shown for you to copy, edit or download as .txt.

Features

  • PDF and images

    Accepts scanned PDFs, document photos and screenshots.

  • Spanish and English

    Recognizes text in Spanish, English or both languages at once.

  • Runs in your browser

    Tesseract.js runs locally — the file is never uploaded to any server.

  • Free, no signup

    No accounts, no watermarks, unlimited uses.

Use cases

  • Digitize scanned documents

    Turn a scanned contract, receipt or record into editable text.

  • Copy text from a photo

    Extract text from a photo of a whiteboard, menu, sign or screenshot.

  • Search scanned PDFs

    Once extracted, you can search the text with Ctrl+F or paste it into another document.

  • Prepare text for translation

    Extract text from a document in one language to paste into a translator.

How is it different?

Most OCR tools upload your file to the cloud. Here everything runs in your browser with Tesseract.js.

  • vs Google Drive / Adobe Scan: Here the file is never uploaded to the cloud. Everything runs in your browser.
  • vs copying by hand: Tesseract recognizes full pages in seconds, with good accuracy for printed text.

Frequently asked questions

Does it recognize handwriting?

Tesseract works well with printed text. Handwriting accuracy is much lower; it's best for typed or printed documents.

How many pages can it process?

No fixed limit, but very long PDFs (over 50 pages) may take several minutes.

How accurate is it?

For printed documents with good resolution, accuracy is above 95%. Blurry photos or very small text will lower it.

Is my file uploaded anywhere?

No. Tesseract.js runs entirely in your browser. The file never leaves your device.

Can I get a searchable PDF?

This tool exports plain text (.txt). To create a PDF with a text layer over the scan you'd need a PDF editor.