There are plenty of languages spoken in India (Hindi, Tamil, Telugu, Gujarati, Marathi, Urdu, Sanskrit, and many others), plus there are many scripts to write on these languages (Devanagari (Nagari), Bengali, Tamil, Perso-Arabic) with regional differences. Billions of people speak these languages, amount of documents that are created on them is enormous. But the complexity of recognition of not very standardized texts and an ease of English alternative creates a very unfortunate situation with no good document management OCR to work with Indian languages.
There are several OCRs that already exist. First of all Tesseract OCR needs to be mentioned. Tesseract is open source OCR tool, that works with Bengali, Hindi, Marathi, Nepali, Sanskrit, Tamil, Telugu and Urdu. However, it is more a good base for your automation project, and it would require a lot of work of turning rather raw OCR capabilities into document management system for your company.
OCRConverter offers a Hindi (Devanagari script) online option. It is Free and easy to use, but it lacks the automation component. You would need a lot of manual labour to integrate it into your document flow, but at least it works.
Our German colleagues have created SanskritReader for Sanskrit. It is a very decent project, but created first of all for academic purposes, and without any automation features.
There are several other online options that you can easily find but most of them look abandoned and we could not really test them.
Main document automation OCR families like ABBYY, IRIS or Kofax do not support Indian languages and are not expressing any desire to include them in any future updates.
The positive side of all that is that Indian OCR field is almost uncovered and steadily growing. And there are many opportunities for talented developers (that India is famous for) to get into growing field of Indian OCR.
What is OCR?
When you scan a document that has text on it, the scanner simply takes a picture of the page and saves it as an image. Text saved in an image cannot be edited or searched, so the data it contains is useless to other software applications. In order to transform this information into an editable format that you can search, copy, and modify without retyping it manually, you need Optical Character Recognition (OCR) software.