Extract text from images, screenshots, and scanned documents. Supports 100+ languages. Runs entirely in your browser.
Language
Upload Image
Drag & drop an image here, or click to upload
JPEG, PNG, WebP, BMP, GIF, TIFF
Extracted Text
Three steps. Works with any image.
This tool uses Tesseract.js — a WebAssembly port of Google's Tesseract OCR engine. When you upload an image, the engine loads language-specific trained data (about 3-15MB per language), then runs optical character recognition directly in your browser. The engine identifies text regions in the image, segments individual characters, and uses neural network pattern matching to convert pixel patterns into text. It returns the recognized text along with a confidence score. Your image is never uploaded to any server — everything runs on your device.