Free Local OCR Engine
Extract text from images · Visual Text Overlay · Select text on pictures · Multiple languages
What does this tool do? Upload any image containing text (a screenshot, photo of a document, or
scanned page) and this tool will read the text for you. The magic? You can actually select
and copy the text directly on the image — just like selecting text on a web page.
Zero uploads — OCR runs entirely in your browser
- Drop Zone: Upload an image containing text
- Visual Overlay: Text appears as a selectable layer right on top of the image
- Languages: Supports English, Spanish, French, German, and more
- Export: Copy as plain text or download a JSON with word positions
View Technical Specifications (For Developers)
- Engine: Tesseract.js v5+ with Web Worker isolation for non-blocking recognition
- Visual Text Overlay: Renders positioned
<span>elements using bounding-box coordinates from Tesseract, creating a transparent selectable layer over the source image - Language Models: Downloaded on-demand (~15MB for English). Cached in browser for subsequent uses
- Output Formats: Plain text, JSON with bounding boxes
(
{text, bbox: {x0, y0, x1, y1}, confidence}) - Performance: Processing offloaded to Web Worker. Speed badge shows total recognition time
Drop an image here for OCR
Supports JPEG, PNG, WebP, BMP, GIF
Fuel the Project
Zero-server OCR. Your images never leave your browser.