Image to Text (OCR)
Pull editable text out of a photo, scan or screenshot — right in your browser. Your image never leaves your device. Then copy or download the result.
Drop an image here
JPG, PNG or screenshot — it stays on your device
Select imageHow to extract text from an image
Add an image
Drop in a photo, scan or screenshot that contains text.
Run OCR
Pick the language and start recognition — it runs on your device.
Copy or save
Edit the result if needed, then copy it or download a .txt file.
Private OCR, no upload
Optical Character Recognition (OCR) turns the text inside an image into characters you can select, search and edit. Most free OCR sites upload your image to a server to do this — a problem when the image is an ID, an invoice, a contract or a private screenshot.
This tool runs the entire OCR engine inside your browser using WebAssembly. Your image is never transmitted. The first time you use a language, a one-time recognition data file downloads and is cached — after that it even works offline.
Frequently asked questions
Is my image uploaded?
No. Recognition happens entirely in your browser — you can confirm there are no image uploads in the network tab. Only a one-time language data file is downloaded.
How accurate is it?
Accuracy is best on clear, high-contrast text. Sharp scans and screenshots work well; blurry photos or unusual fonts are harder, as with any OCR.
Which languages can it read?
English plus many others in the dropdown. Each language loads its own data file the first time you select it.
Does it keep the layout?
It extracts the text content. Complex multi-column layouts may be reordered; simple text and paragraphs come through cleanly.