Hindi characters, recognized properly
The OCR reads Devanagari with its vowel signs (matras), conjunct consonants and the nukta, as well as the English words and numbers common in Indian business documents. Other Devanagari languages such as Marathi and Nepali are recognized too.
Handwritten Hindi is not supported yet. Printed and typewritten Hindi work well.
No language setting needed
You don't pick a language. The OCR detects it on each page, so a folder of Hindi files and English files can go in the same batch.
Indian documents often switch between Hindi and English on the same page. The OCR reads both without you choosing a language.
Searching Hindi text in the result
Some PDF viewers handle Devanagari better than others when searching and copying. If copied text looks jumbled, with vowel signs in the wrong place, try another viewer before assuming the OCR is wrong.
Hindi documents people convert
- Invoices and GST bills
- Land and property records
- Court documents and affidavits
- Government forms and certificates
- School and university records
Older and harder documents
Clear printed pages work well. Faded stamp paper, faint photocopies and low-resolution scans cause most errors, so scan at 300 dpi where you can.
If a file already has a text layer, it is returned unchanged. Tick Force OCR to replace a poor existing text layer; see how Force OCR works.
How to convert
- Create an account. It comes with 10 free pages to try it, valid for 7 days, with no credit card.
- Drop in PDFs, JPGs or PNGs, up to 50 files at a time.
- Most files are ready in 10 to 30 seconds. Download each searchable PDF, or all of them as a ZIP.
After that, monthly plans start at $9 for 250 pages, or you can buy a one-off pack of 250 pages for $14.99 or 1,000 pages for $39.99. Pack pages never expire.
Questions
- Do I need to choose Hindi before uploading?
- No. The language is detected automatically, including on pages that mix Hindi and English.
- Does it read handwritten Hindi?
- Not yet. Printed and typewritten Hindi is fully supported.
- Will the document look different?
- No. The page image stays the same. OCR adds an invisible text layer on top of it.
- Are my files kept?
- No. Files are deleted automatically within an hour of processing.