फोटो वा PDF यहाँ तान्नुहोस् वा क्लिक गरेर छान्नुहोस्
JPG · PNG · WEBP · BMP · PDF — स्क्रिनसट सिधै पेस्ट (Ctrl+V) पनि गर्न सकिन्छ
Extract Nepali text from any photo, screenshot, scanned document or PDF — free, instantly, and without your file ever leaving your device. Upload a picture of a Nepali book page, citizenship document, newspaper cutting, handwritten-style printed notice or old office letter, and our OCR (Optical Character Recognition) engine reads the देवनागरी script and gives you editable Nepali Unicode text you can copy into Word, email, Facebook or anywhere else.
This is the only Nepali OCR tool with a built-in Preeti font pipeline. Old Nepali PDFs made in Preeti, Kantipur or similar legacy fonts contain scrambled letters like g]kfn when you try to copy from them. Paste-copying fails — but drop that PDF here and the tool detects the Preeti encoding automatically and converts it straight to clean Unicode (नेपाल) using the same engine as our Preeti to Unicode converter. Scanned image, digital PDF or legacy-font PDF — one tool handles all three.
1. Drag your photo or PDF into the box above, click it to browse, or simply paste a screenshot with Ctrl+V (Cmd+V on Mac).
2. Choose the document language — नेपाली, English, or both for mixed documents like forms and bills.
3. The text is extracted automatically with a live progress bar. Multi-page PDFs are processed page by page.
4. Edit the result right in the box if needed, then copy it or download it as a .txt file.
Yes — genuinely. Unlike most "image to text" websites that upload your document to their server, this tool runs the entire OCR engine inside your browser using WebAssembly. Your citizenship scan, bank document or office letter is never uploaded, stored or seen by anyone. The first run downloads the Nepali language model once; after that it works at full speed on your own device.
OCR quality depends on image quality. For best results: use a sharp, well-lit photo taken straight-on (not at an angle); prefer scans at 300 DPI; avoid shadows across the page; and crop to just the text if the photo has lots of background. Clean printed text converts almost perfectly — very stylised fonts, low-resolution photos and handwriting are harder for any OCR engine. The extracted text appears in an editable box, so you can fix the odd character before copying.
Once your Nepali text is in Unicode, every tool on this site works with it: continue writing by voice with Nepali Voice Typing, edit and export documents in Nepali Docs, convert it back for old systems with Unicode to Preeti, or type more with English to Nepali typing.
Yes — both kinds. Scanned PDFs are rendered page by page and read with OCR. Digital PDFs with selectable text are read directly, which is faster and 100% accurate — and if that embedded text turns out to be in Preeti font encoding, it is converted to Unicode automatically.
No. The OCR engine (Tesseract, compiled to WebAssembly) runs entirely in your browser. Your file never leaves your device — you can even disconnect from the internet after the page loads and it still works.
Yes — this is our speciality. When a PDF's embedded text looks like Preeti-encoded ASCII (e.g. sf7df8f}+), the tool detects it and converts it to proper Unicode (काठमाडौं) automatically. A button lets you view the original Preeti text too.
Printed and typed text works best. Clear, neat handwriting may partially work, but OCR engines — including Google's — still struggle with Nepali handwriting. For printed books, notices, letters and forms, accuracy is very high.
JPG, PNG, WEBP, BMP and PDF, up to 25 MB per file. You can also paste a screenshot directly from your clipboard with Ctrl+V, or take a photo with your phone camera when browsing on mobile.
On first use your browser downloads the Nepali language model (a few MB) once. It is cached afterwards, so every later conversion starts instantly.