Image → editable Urdu

Urdu OCR — Image to Urdu Text

Choose a clear Urdu screenshot or printed image and extract editable Unicode text. The OCR runs in your browser so you can review, clean and continue writing without sending the image to a WriteUrdu OCR server.

Extract Urdu from an image

Best for clear screenshots and printed Urdu. OCR is probabilistic, so review important names, numbers and punctuation.

Choose an image
Source image PNG, JPEG or WebP · up to 12 MB
The first OCR run downloads the Tesseract browser engine and Urdu language data from a pinned CDN. The selected image itself is not uploaded by WriteUrdu.
Review before use

Accuracy note: clear screenshots and ordinary printed Urdu are the target for this first version. Complex Nastaliq layouts, handwriting, newspapers, skewed photos and mixed English may need more correction.

What Urdu OCR does

OCR means optical character recognition: software looks at pixels in an image and predicts the text they represent. This tool uses an Urdu recognition model in the browser and returns ordinary Unicode text that you can edit, search, copy and export.

Good inputs

Use a tightly cropped screenshot or scan with readable Urdu, strong contrast and enough resolution. Clear printed text generally gives OCR more information than a distant photo, decorative background or heavily compressed social image.

Why the result still needs review

Urdu OCR can confuse similar letter shapes, dots, punctuation and mixed-script content. Nastaliq also has complex joining and layout. Treat OCR as a fast starting point, not a guarantee that every character is correct.

Continue with WriteUrdu

After extraction you can send the result to the Urdu Text Cleaner & RTL Fixer or open it directly in the WriteUrdu editor. The handoff uses session-only browser storage rather than putting your text in a URL.