How to convert an image to text
- Drop a photo, screenshot or scan on the box, tap to choose a file, or paste a screenshot straight from your clipboard with Ctrl+V (Cmd+V on a Mac).
- Choose the language of the text. Add a second language for mixed documents, such as English and French.
- Press Recognise text. The first run downloads the OCR engine and language data (a few megabytes), which your browser then caches.
- Edit the result if needed, then copy it or download it as a text file.
What is OCR?
Optical character recognition turns pictures of text into real, editable text. It saves retyping receipts, letters, book pages, slides, whiteboard photos, screenshots of error messages and printed forms. This tool uses Tesseract, the open-source OCR engine originally developed at HP and later maintained by Google, compiled to run in your browser, with support for more than 30 languages including Chinese, Japanese, Arabic and Hindi.
Tips for accurate results
- Use a sharp, well-lit, straight image. Photograph pages flat, without shadows or strong angles.
- Higher resolution helps small print: aim for text at least 10 pixels tall.
- Crop away backgrounds, photos and decorative elements.
- Handwriting is not recognised reliably; OCR is designed for printed text.
Unlike most OCR websites, your image is never uploaded: recognition happens on your device, so it is safe for IDs, medical letters and confidential papers.
Can it read handwriting?
Neat block capitals sometimes work, but cursive handwriting is not supported. Tesseract is trained on printed text.
Can I OCR a scanned PDF?
Yes. Drop the PDF; each page is rendered and recognised. Tick "Also create a searchable PDF" to get a PDF with selectable text.
Why are some characters wrong?
Blurry, low-resolution or skewed images cause errors such as 0 for O or 1 for l. Retake the photo more sharply or crop tighter.