Image to Text Converter
Extract Nepali and English text from photos, screenshots, scanned documents and PDFs. Recognition runs on your device, so your files are never uploaded.
Tips for better results
Choose the right language
Choose नेपाली or English alone when the image has only one language. It is faster and more accurate.
Use clear images
Straight, well-lit photos with sharp text work best. Crop out backgrounds and avoid shadows or blur.
Check and edit
OCR can make mistakes. The text box is editable, so fix any errors before you copy or download.
Nepali OCR that runs in your browser
OCR (optical character recognition) turns a picture of text into real text you can edit, search and copy. This tool uses Tesseract, a widely used open-source OCR engine, with language models for Nepali (Devanagari) and English. It runs entirely in your browser, so your documents stay on your device.
Devanagari is harder to read than English: letters are joined by the headline (शिरोरेखा), matras sit above and below the letters, and conjuncts merge several letters into one shape. For the best results, use a sharp, well-lit, straight photo, crop it to the text, and choose only the language the image actually contains.
The result is standard Unicode Nepali, ready for Facebook, Google Docs or email. If you need it in Preeti for a document or printing press, use Copy as Preeti, which converts it with our Unicode to Preeti converter.
OCR is never perfect, so always proofread important text. Pay particular attention to ि/ी, ु/ू, ँ/ं and numbers. Our guide to Nepali OCR has detailed tips and a proofreading checklist.
Frequently asked questions
Does it read Nepali (Devanagari) text?
Yes. Choose नेपाली to read printed Nepali in Unicode. Use Copy as Preeti to get the result in Preeti for Word or PageMaker. Handwriting is not supported.
Why is the first conversion slow?
The first time, your browser downloads the OCR engine and language data (a few MB). It is saved in your browser, so later conversions start much faster.
Can it read PDFs?
Yes. Each page of the PDF is read in order, and the text is labelled by page number.
Are my images uploaded?
No. Text recognition runs in your browser. Only the OCR engine and language data are downloaded. Your files never leave your device.