SnapOCR

Runs in your browser. Nothing is uploaded.

Extract text from images. Your files never leave your device.

Drop in a screenshot, photo or scan. SnapOCR reads it on your own computer and puts the text on your clipboard. Free, with no account and no upload.

Drop an image here

or

JPG, PNG, GIF, WebP or BMP, up to 10 MB

Reading your image…

Extracted text

Download as

How it works

  1. Add an image

    Drag a file onto the page or choose one from your device. Screenshots, photos of documents and scanned pages all work.

  2. Text is recognised on your device

    Recognition runs inside your browser using Tesseract, an open-source OCR engine. The image is never sent to a server.

  3. Copy or download

    The text is placed on your clipboard automatically. Download it as a plain text, Markdown, Word-compatible or data file if you prefer.

Getting the best results

Works best with

  • Sharp, well-lit images with dark text on a light background
  • Screenshots, scanned pages and PDF pages saved as images
  • Text that is straight, not rotated or photographed at an angle
  • Printed English text. Other languages and handwriting are not yet supported.

Download formats

  • TXTPlain text
  • MDMarkdown
  • HTMLStandalone web page
  • RTFOpens in Word and Pages
  • CSVLines, words or statistics
  • JSONStructured, with word details
  • XMLStructured document

Private by design

Most online OCR tools upload your image to their servers. SnapOCR does not. The recognition engine is downloaded to your browser once and every image is processed locally, so statements, contracts and personal documents stay on your machine.

The only outside service is Fathom, a privacy-focused analytics tool that counts visits without cookies or personal data.

Copied to clipboard