Kalkuly
EN
Image to text (OCR)
Turn a photo, a screenshot or a scanned page into editable text. Recognition runs entirely in your browser: your image is never uploaded to any server.
How to use it
Drop an image, click to choose one or paste a screenshot with Ctrl+V. Choose the language of the text, then copy the result. The first use downloads the recognition engine (a few seconds).
Formula and method
Optical character recognition (OCR) with Tesseract, an open-source engine originally developed by HP and maintained by Google. It finds lines of text in the image, then recognises each word with a neural network (LSTM) trained for the chosen language.
Worked example
Photograph a page of a book, drop the photo here and select English: after a few seconds the text appears, ready to paste into Word or Google Docs. For best results, use a sharp, well-lit image where the text is straight.
Frequently asked questions
Is my image sent to a server?
No. The OCR engine runs in your browser with WebAssembly, so your images and the extracted text stay on your device.
Can it read handwriting?
Printed text works best. Neat handwriting is sometimes recognised, but results are much less reliable.