Image to Text
Pull the words out of a screenshot, scan or photo and get editable text back. The OCR runs inside your browser, so the image is never uploaded.
Images never leave your device
How to convert an image to text
Choose a JPG, PNG or WebP file, pick the language the text is written in, and press Extract text. The first run downloads the recognition engine, which takes a few seconds on a normal connection; after that it is cached and later images are much quicker. When it finishes you get editable text you can correct, copy, or download as a .txt file.
How to get the most accurate result
Optical character recognition reads shapes, so the sharper and straighter the text, the better it does. Screenshots and flat scans usually come out near perfect. For a photo of a page, fill the frame with the text, hold the camera square rather than at an angle, and avoid shadows falling across the writing. If part of a photo has the text you need, crop to that part first and the result improves noticeably. Very small text is worth enlarging before you run it.
What it handles well, and what it does not
Printed text in clear type is its strength: screenshots, receipts, book and textbook pages, forms, slides, signboards and PDF pages you have exported as images. Handwriting is not reliable, and neither is heavily stylised lettering, text over busy photographs, or very low-resolution captures. Tables and multi-column layouts come back as plain lines rather than preserved columns, so expect to tidy the order by hand.
Your image stays on your device
Unlike most online OCR services, nothing here is uploaded. The recognition engine is downloaded to your browser and does the work locally, which is why the first run needs a moment. That means you can safely read text from documents you would not want to send to somebody else's server.
More image tools
Frequently asked questions
Is this image to text converter really free?
Yes, and there is no sign-up, no watermark and no page limit. The recognition runs on your own device, so there is no server cost to pass on.
Does my image get uploaded to a server?
No. The OCR engine is downloaded into your browser and the image is read locally on your device. Nothing is sent anywhere, which is why it works even if you go offline after the first run.
Which languages can it read?
English, Hindi in Devanagari script, and Tamil. You can also pick a combined option such as English plus Hindi when a page mixes both, which is common on Indian forms and signage.
Why is the first extraction slower than the rest?
The first run downloads the recognition engine and the language data, around 5 MB. Your browser caches it, so every image after that starts almost immediately.
Can it read handwriting?
Not reliably. The engine is trained on printed type, so neat printed capitals sometimes work but ordinary handwriting usually does not. Printed text, screenshots and scans are what it is good at.
Why did it get some characters wrong?
Accuracy depends on how clearly the text is captured. Blur, angle, shadow, low resolution and decorative fonts all hurt it. Crop to the text, straighten the shot and try again. The extracted text is editable so you can fix anything it missed.
Can I extract text from a PDF?
Convert the PDF to images first with our PDF to Images tool, then run each page through this one.