Image to Text Converter
Extract text from an image, screenshot or scanned page (OCR) in English, Urdu, Arabic, Hindi and 12 more languages.
The picture is read in your browser and never uploaded.
Drag & drop an image here
or click to choose an image; you can also paste a screenshot with Ctrl+V
Extracted text
Related Image Tools
This image to text converter uses OCR (optical character recognition) to pull the words out of a picture: a screenshot, a photo of a page, a scanned document, a receipt or a slide. The text appears in an editable box, ready to copy into an email or document or save as a text file. It reads English, Urdu, Arabic, Hindi, Bengali, Chinese, Japanese, Korean and most European languages, and the picture stays on your device.
How to Convert an Image to Text
- Choose the language the text is written in. For a page that mixes Urdu or Hindi with English, choose the combined option.
- Add the picture, or copy a screenshot and paste it on the page with Ctrl+V (Cmd+V on a Mac).
- Press Extract text. The first run downloads the engine; later runs start straight away.
- Check and copy. Correct any mistakes in the box, then copy the text or download it as .txt.
Getting the Most Accurate Result
- Pick the right language. The engine expects the alphabet and words of the language you choose; the wrong one gives nonsense.
- Crop to the text. Logos, photos and table lines around the text can be read as stray characters. Trim them with the Image Cropper first.
- Shoot pages flat and straight. Hold the phone parallel to the page, in good light, without shadows. Rotate a sideways photo with the Image Rotator.
- Small text is enlarged for you. Pictures under 1200 pixels are scaled up before reading, which helps with cropped screenshots.
What OCR Can and Cannot Read
| Source | Typical result |
|---|---|
| Screenshots of apps, websites and documents | Very accurate |
| Flat scans and clear photos of printed pages | Accurate; check numbers and names |
| Receipts and faded print | Mixed; small or faint text is often misread |
| Urdu Nastaliq, calligraphy, decorative fonts | Often inaccurate |
| Handwriting | Poor, except very neat print |
How It Works
The tool runs Tesseract, the open-source OCR engine originally developed at HP and later maintained with Google's support, compiled to run inside the browser (Tesseract.js). The engine and the language data are downloaded from the jsDelivr network the first time and kept in your browser's storage. Your picture is processed on your own device. To turn several pictures into one document instead, use Image to PDF; note that the PDF it makes is not searchable.
Frequently Asked Questions
How do I extract text from an image?
Add the image (or paste a screenshot with Ctrl+V), choose the language the text is written in, and press Extract text. The text appears on the right, where you can correct it, copy it or save it as a .txt file.
Can it read Urdu and Arabic?
Yes. Choose Urdu, Arabic, or Urdu + English for text that mixes both. Printed Naskh text is read best; Nastaliq calligraphy and decorative fonts give less accurate results, so check the text before using it.
Can it read handwriting?
Only very neat, printed-style handwriting. The engine is trained on printed text, so joined or quick handwriting mostly comes out wrong.
How accurate is the text?
On a clear screenshot or a flat, well-lit scan, usually nearly perfect. Accuracy drops with blur, low resolution, curved pages, shadows, coloured backgrounds and unusual fonts. The confidence figure under the text is the engine's own estimate.
Why does it take a while the first time?
The first time, your browser downloads the recognition engine and the language data, a few megabytes for most languages. They are stored in your browser, so the next pictures are read much faster.
Is my image uploaded?
No. The text is read on your own device. Only the engine and language files are downloaded, from the jsDelivr network; your picture and its text are never sent anywhere.