Image to Text (OCR)

Extract text from a photo or scanned image, right in your browser.

About this tool

Extracts text from a photo, screenshot, or scanned image — printed text only, not handwriting — using the open-source Tesseract OCR engine compiled to run directly in your browser. Nothing is uploaded anywhere; recognition happens entirely on your device.

For a scanned PDF instead of an image, convert it to JPG first with PDF to JPG, then run each page through this tool.

Frequently asked questions

Why did it take a while before anything happened?
The OCR engine and its language data (a few megabytes) are only downloaded the first time you use this tool, not when the page first loads — this keeps the page itself fast, at the cost of a short delay before your first extraction.
Does this work on handwriting?
Not reliably — this engine is trained for printed and typed text. Handwriting recognition is a much harder problem and generally needs a different, specialized model.
Why is some of the text wrong or missing?
Accuracy depends heavily on image quality — low resolution, blur, skewed angles, low contrast, or unusual fonts all reduce accuracy. A sharp, well-lit, straight-on photo of clear printed text works best.
Only English?
Yes, for now — this tool ships with English-language recognition data only.
Is my image uploaded anywhere?
No — recognition runs entirely on your device using WebAssembly. Nothing is sent to any server.