Copy the text out of a photo or screenshot — read right in your browser.
Processed on your device
Your files are processed in your browser and never uploaded.
How do I copy text from an image or screenshot?
Open Image to Text on Tools-Bench, drop in the picture or paste a screenshot, pick the text's language and click Read text. The words appear on the page, ready to correct, copy or save as .txt. Recognition runs in your browser, so the image is never uploaded. It reads printed and on-screen text, not handwriting.
About Image to Text
Image to Text uses OCR (optical character recognition) to read the words in a picture — a screenshot of an error message, a photo of a receipt, a slide, a page from a book, a sign — and gives them back as plain text you can copy, edit or save. The recognition engine (Tesseract) runs in your browser: it downloads once, roughly 15 MB, and is cached afterwards, so your images are never uploaded and there's no daily limit. It reads printed and on-screen text in 10 languages; handwriting is mostly beyond it.
How to use Image to Text
1
Add an image
A screenshot, a photo taken with your phone or a scanned page — PNG, JPG, WebP and iPhone HEIC all work. You can also paste a screenshot with Ctrl+V (⌘V on a Mac).
2
Pick the text's language
It tells the engine which letters and words to expect; choose the language the text is written in.
3
Read the text
The words appear on the page. Fix anything that was misread, then copy them or download a .txt file.
Frequently asked questions
Can it read handwriting?
Not reliably. The engine is trained on printed and on-screen text: neat block capitals sometimes work, joined-up handwriting almost never does. For printed text, accuracy is high when the image is sharp.
How do I get the best results?
Use the sharpest, straightest image you have: crop to the text, avoid glare and shadows, and make sure the letters aren't tiny. Screenshots usually read almost perfectly because the text is clean.
Is there a limit on how many images I can read?
No. Recognition runs on your own device, so there's no daily quota, no sign-up and no watermark. The first run downloads the engine (roughly 15 MB), and each extra language downloads its own model once.
What about a scanned PDF?
Use OCR — Searchable PDF for that: it reads every page of a scanned PDF and can also turn it into a searchable PDF.