How it works
Text recognition (OCR — Optical Character Recognition) runs entirely in your browser using Tesseract.js, a WebAssembly build of the open-source Tesseract OCR engine (the same one used by Google in other products). No image is ever uploaded to a server — processing happens locally, which also means it works fine with sensitive photos, screenshots and scans without privacy concerns.
Tips for better results
Sharp, well-lit images with horizontally aligned text tend to give much more accurate results. Select the correct language of the text before extracting — it significantly improves recognition accuracy.
Why the first extraction takes longer
The first time, your browser downloads and initializes the OCR engine (a few megabytes) and the chosen language data; later extractions in the same session tend to be faster.
Frequently asked questions
Is my image uploaded to a server?
No — all OCR processing happens locally in your browser using WebAssembly. Your image never leaves your computer.
Which languages are supported?
Portuguese, English and Spanish. Select the correct language of the text in the image before extracting for better accuracy.
Why did it take so long to process?
OCR text recognition is computationally heavy, even more so running in a browser. Large images or ones with a lot of text can take a few seconds to over a minute.
Does it work with handwritten text?
Tesseract is optimized for printed/typed text. Handwritten text can have a much lower accuracy rate.