Image to Text (OCR)
Extract text from a photo, screenshot, or scanned document. The recognition runs entirely in your browser — the image is never uploaded anywhere.
Click to upload or drag and drop
JPG, PNG, or WebP
How it works
This tool uses an open-source OCR (optical character recognition) engine that runs as WebAssembly directly in your browser tab. Your image is read and processed locally — it's never sent to a server. The first run downloads the recognition engine and language data (a few megabytes), which your browser then caches for future use. You can choose from six languages (English, Spanish, French, German, Hindi or Portuguese), and the result comes with a confidence score.
This tool reads image files such as photos, screenshots and single-page scans. For a PDF, use Scanned PDF to Text (OCR) if its pages are scans, or the PDF to Text Converter if it already has selectable text.
Getting the best results
- Clear, well-lit photos and screenshots work best — aim for at least 200 DPI equivalent
- Straight, non-rotated text is recognized more reliably than skewed or curved text
- Printed text is much more reliable than handwriting
- Code, equations, and unusual symbols are recognized poorly — this works best on ordinary prose
Frequently asked questions
Is my image uploaded anywhere?
No — the recognition engine runs as WebAssembly directly in your browser tab. Your image never leaves your device.
Why does the first run take longer?
The OCR engine and language data (a few megabytes) are downloaded on first use and cached by your browser, so later runs start much faster.
Does this work on handwriting?
It's built for printed text and works poorly on handwriting, code, or heavily stylized fonts.
Can it read scanned PDFs?
Not in this tool, which takes image files (JPG, PNG, WebP). For a scanned PDF use the Scanned PDF to Text (OCR) tool; for a PDF that already has selectable text, use the PDF to Text Converter.