Drop an Urdu image and the text comes out as Unicode Urdu you can paste into Word, InPage-free documents, WhatsApp or Facebook. It is shown in a Nastaliq font, and the Word file is set to right to left with Urdu as the language, so it opens the right way round.
Nothing is uploaded. The image is read by a model that runs inside your browser, so it works for private documents and keeps working on a slow connection once the model is cached.
Nastaliq is hard. Here are our real numbers.
Most Urdu print, newspapers and InPage documents use Nastaliq: letters join on a slope and stack, which no small OCR model reads perfectly. We tested on Urdu rendered in Noto Nastaliq Urdu and, separately, in the simpler Naskh style used on many websites and signs.
The default engine (Tesseract 5) got 71% of characters right on Nastaliq, and 95.7% on Naskh. Nastaliq results need checking: expect a few wrong or missing letters per line, especially dots (ب پ ت ث) and small letters at the end of words. Naskh and web screenshots come out much cleaner.
We would rather show these numbers than claim “99% accurate”. If you have InPage files, export them as text instead; OCR is for images.
How to do it
- Make sure Urdu · اردو is selected.
- Drop, paste or photograph the page. Larger text reads better: zoom in before you take a screenshot.
- Check the text, then copy or download Word.
اردو او سی آر: تصویر سے اردو متن
کسی بھی اردو تصویر، اسکین یا اسکرین شاٹ سے یونی کوڈ اردو متن حاصل کریں، جسے آپ ورڈ، واٹس ایپ یا فیس بک میں براہِ راست استعمال کر سکیں۔ ورڈ فائل دائیں سے بائیں کھلتی ہے۔
آپ کی تصویر کہیں اپ لوڈ نہیں ہوتی۔ پڑھنے والا ماڈل ایک بار ڈاؤن لوڈ ہو کر آپ کے براؤزر میں ہی چلتا ہے۔
ایمانداری سے: نستعلیق پڑھنا مشکل ہے۔ ہماری جانچ میں نستعلیق کے 71% حروف درست پڑھے گئے، اور نسخ کے 95.7%۔ استعمال سے پہلے متن ایک بار ضرور پڑھ لیں۔
- اوپر زبان میں "Urdu · اردو" منتخب کریں۔
- تصویر ڈالیں، اسکرین شاٹ پیسٹ کریں یا کیمرا دبائیں۔
- متن جانچیں، پھر کاپی کریں یا ورڈ فائل ڈاؤن لوڈ کریں۔
How accurate it is (measured, not claimed)
We render known text in each script as images in five styles: a clean screenshot, small 15 px text, a photo (rotated, soft focus, uneven light, JPEG), a dark-mode screenshot and a rough photo (low contrast, blurrier, more tilt, heavy JPEG). Each engine reads them in a browser, and we count the characters it gets right (100% minus the character error rate). The default engine for each script is the one with fewer character errors (within half a point, the one with more whole words right).
| Script | Engine | All | clean | small | photo | dark | rough | s / image |
|---|---|---|---|---|---|---|---|---|
| Urdu (Nastaliq) | Tesseract 5default | 70.7% | 77.1% | 76.7% | 68.2% | 69.1% | 62.9% | 0.2 |
| Urdu (Nastaliq) | PaddleOCR PP-OCRv5 | 30.7% | 44.7% | 32.3% | 25.6% | 44.2% | 7.7% | 0.9 |
| Urdu (Naskh) | Tesseract 5default | 95.7% | 97.1% | 97.0% | 97.4% | 96.4% | 90.6% | 0.4 |
| Urdu (Naskh) | PaddleOCR PP-OCRv5 | 81.8% | 90.0% | 82.0% | 84.6% | 86.7% | 65.3% | 1.8 |
Measured 2026-11-03 on Intel(R) Core(TM) i7-14700KF (28 threads), headless Chromium, no WebGPU adapter, cross-origin isolated. Real photos have not been validated for this tool. Generated images are easier than real photos of real paper, so treat these as a ceiling for printed text, not a promise. Tesseract was partly trained on font-rendered text, which may flatter it on generated images. The test text and the script that draws the images are in the project repository.
Private by design
The image is read by an open-source model that runs in your browser (PaddleOCR or Tesseract, both Apache-2.0). It downloads once, about 12 MB for most scripts, and is cached for next time. Your images, the text and your downloads stay on your device: there is no upload, no account and no log of what you read.
Questions people ask
Are my images uploaded to a server?
No. The reading model is downloaded to your browser once and runs on your own phone or computer. The image, the text and the files you download never leave your device. You can switch off your internet after the model has loaded and it still works.
Is it free? Is there a limit?
It is free, with no sign-up, no watermark and no daily cap. Because your own device does the work, there is no server cost for us to recover. Add as many images as you like; they are read one after another.
Which file types can I use?
PNG, JPG, WebP, GIF and BMP images, screenshots pasted with Ctrl+V, and photos from your camera. For a PDF, take a screenshot of the page or export it as an image first.
Does it read Nastaliq?
Yes, but less accurately than Naskh or English: 71% of characters on our Nastaliq test images. Read the result through before you use it. For Naskh text it is 95.7%.
Will the Word file be right to left?
Yes. Paragraphs are set to right to left and marked as Urdu, and the text uses a Nastaliq font where you have one installed.
Can it read Urdu mixed with English?
It reads the Urdu. English words inside an Urdu line may come out wrong; for a mostly-English page choose English instead.
Does it read Urdu handwriting?
No. Handwritten Urdu is beyond the models that can run in a browser today.