Text Extractor (OCR)
Runs on your device. Nothing is sent.Your image never leaves your device - the OCR engine runs right here in the page.
Drop an image here, click to choose one, or paste a screenshot with Ctrl+V.
Languages in the imagePick up to 3. Fewer languages read faster and more accurately.
Add an image to get started.
Pull text out of photos, screenshots and scans in 17 languages - recognised entirely in your browser.
This tool turns text in an image back into text you can copy, edit and search - receipts, screenshots, book pages, signs, scanned letters. It uses Tesseract, the open-source OCR engine, compiled to WebAssembly so it runs inside your browser tab. The engine and the language models are served from Gizke.com itself, so neither your image nor even the fact that you're running OCR is shared with anyone.
Frequently asked questions
Is my image uploaded anywhere?
No. Recognition happens entirely on your device. The page downloads the OCR engine and the language models you pick from Gizke.com the first time, then your browser keeps them - your image itself is never sent anywhere.
How do I get the best results?
Use a sharp, well-lit image where the text is at least a dozen pixels tall, straight rather than angled, and select only the languages that actually appear. Screenshots and scans usually read almost perfectly; photos of curved pages or handwriting are much harder.
Can it read handwriting?
Only neat block letters, and not reliably. Tesseract is built for printed text; for handwriting expect partial results.
Why is the first run slower?
The first time, your browser downloads the engine (about 4 MB) and each selected language model (about 0.5-3 MB). They're cached afterwards, so later runs start almost instantly - and keep working offline.