IndicPDF Translation (beta)

What it does

• Between English and each of the Indian languages IndicPDF reads (Hindi, Telugu, Tamil, Kannada, Malayalam, Marathi, Bengali, Gujarati, Punjabi, Odia, Assamese and Sanskrit), and between any two of them. Telugu and Hindi have been read by people who know them; the others are checked by machine tests so far.
• Urdu, in beta: English to Urdu and Urdu to English, written right to left. Nobody who reads Urdu has reviewed a translation yet.
• Also in beta, English to and from Nepali, Konkani, Maithili, Dogri, Bodo, Sindhi (Arabic script), Kashmiri, Manipuri (Meetei Mayek) and Santali (Ol Chiki). Nobody who reads them has reviewed a translation yet.
• It takes a Word file or a PDF, and returns a Word file with the same layout.
• A scanned PDF, or a photo of a page (JPG, PNG, WebP, GIF, BMP, TIFF), is read first by OCR, and the reading is worth checking, as it is with the OCR tool. A photo comes back as a clean Word file rather than a copy of the original layout.
• Numbers, dates, section references, web addresses and names are kept exactly as written.
• A sentence that could not be translated acceptably stays in the original language and is highlighted.
• You need an account with a confirmed e-mail address, and a limited number of pages a day.
• The text is translated by an outside machine-translation service. What is sent, and what is not, is on the Privacy page.

More: English to Telugu, English to Hindi, Telugu to English, Hindi to English and how the layout is kept.

English and: Tamil, Marathi, Kannada, Malayalam, Bengali, Gujarati, Punjabi, Odia, Assamese, Sanskrit, Urdu, each both ways.