This tool reads the text in a picture and gives it back to you as plain text you can copy, edit or save. It works on screenshots, photos of printed pages, scanned forms and receipts. The technical name is OCR (optical character recognition).
Everything happens in your browser. The image never leaves your device, which matters when the picture is an ID card, a bank statement or a private message.
How to use
- Click Choose an image, drop it here, or paste it. You can also copy an image and press Ctrl+V anywhere on this page.
- Pick the Language of the text. Use English + Urdu for pages that mix both scripts.
- Click Extract text and wait for the progress bar. The first run downloads the OCR engine and the language data, so it takes longer.
- Check the result in the Extracted text box and fix any wrong letters right there.
- Click Copy text or Download .txt.
What makes OCR accurate (and what ruins it)
OCR software looks for the shapes of letters. Anything that blurs or warps those shapes lowers accuracy, so a few minutes spent on the photo usually saves more time than fixing errors afterwards.
- Sharpness. Hold the phone steady and tap to focus on the text. A blurry photo of a page is the most common reason for garbage output.
- Size of the text. Tiny text in a huge photo is hard to read. Crop to the part you need before you upload. Small screenshots are enlarged automatically before reading.
- Straight lines. Shoot the page from directly above, not at an angle. Curved lines near a book's spine often come out jumbled.
- Contrast. Dark text on a plain light background works best. Shadows across the page, glare on glossy paper and coloured backgrounds all cause mistakes.
- The right language. Picking English for an Urdu page gives nonsense, because the engine tries to match English letter shapes.
Handwriting is the hard case. Neat block capitals sometimes work, but joined handwriting mostly will not. That is a limit of the engine used here (Tesseract), not something a setting can fix.
Urdu, Arabic and Hindi
Right-to-left scripts work, and the result box switches direction so the text reads correctly. Urdu printed in Nastaliq (the slanted style used in most newspapers and books) is harder to recognise than Naskh, so expect more errors and keep the image large and sharp. For mixed pages, such as a Pakistani form with English labels and Urdu answers, the English + Urdu option reads both, though it is slower than a single language.
Why the first run is slow
The OCR engine and each language's data file are a few MB. They download once the first time you use a language, then your browser keeps them, so later runs start in a second or two. After that, the reading itself runs on your own device.
FAQ
Is this image to text converter free, and are my images uploaded?
It is free with no sign-up and no limit on images. The OCR runs inside your browser, so your pictures are never uploaded to a server. Only the engine and language files are downloaded the first time.
Can it read text from a screenshot on my phone?
Yes. Screenshots are often the easiest case because the text is already sharp and straight. On a phone, choose the screenshot from your gallery with the upload box.
Why is the extracted text full of strange characters?
Usually the language setting does not match the text, or the photo is blurry, tilted or too small. Pick the right language, crop to the text area and try a sharper photo.
Does it keep the formatting, like tables and bold text?
No. You get plain text with line breaks. Tables come out as lines of words, so you may need to tidy them after pasting into a spreadsheet or document.