Skip to content
Transliteration package for Indian scripts
Find file
New pull request
Pull request Compare This branch is 6 commits behind sanskrit:master.
Fetching latest commit...
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.



Sanscript is a transliteration library for Indian languages. It supports the most popular Indian scripts and several different romanization schemes. Although Sanscript focuses on Sanskrit transliteration, it has partial support for other languages and is easy to extend.


Sanscript is simple to use:

var output = Sanscript.t(input, from, to);

Here, from and to are the names of different schemes. In Sanscript, the word "scheme" refers to both scripts and romanizations. These schemes are of two types:

  1. Brahmic schemes, which are abugidas. All Indian scripts are Brahmic schemes.
  2. Roman schemes, which are alphabets. All romanizations are Roman schemes.

By default, Sanscript supports the following Brahmic schemes:

  • bengali
  • devanagari
  • gujarati
  • gurmukhi
  • kannada
  • malayalam
  • oriya
  • tamil
  • telugu

and the following Roman schemes:

  • hk (Harvard-Kyoto)
  • iast (International Alphabet of Sanskrit Transliteration)
  • itrans (ITRANS)
  • itrans_dravidian (ITRANS with support for Dravidian short "e" and "o")
  • kolkata (National Library at Kolkata)
  • slp1 (Sanskrit Library Phonetic Basic)
  • velthuis (Velthuis)
  • wx (WX)

Disabling transliteration

When Sanscript sees the token ##, it toggles the transliteration state:

Sanscript.t('ga##Na##pa##te', 'hk', 'devanagari'); // गNaपte
Sanscript.t('ध##र्म##क्षेत्रे', 'devanagari', 'hk'); // dhaर्मkSetre

When Sanscript sees the token \, it disables transliteration on the character that immediately follows. \ is used for ITRANS compatibility; we recommend always using ## instead.

Sanscript.t('a \\a', 'itrans', 'devanagari'); // अ a
Sanscript.t('\\##aham', 'itrans', 'devanagari'); // ##अहम्

Transliterating to lossy schemes

A lossy scheme does not have the letters needed to support lossless translation. For example, Bengali is a lossy scheme because it uses for both ba and va. In future releases, Sanscript might let you choose how to handle lossiness. For the time being, it makes some fairly bad hard-coded assumptions. Corrections and advice are always welcome.

Transliteration options

You can tweak the transliteration function by passing an options object:

var output = Sanscript.t(input, from, to, options);

options maps options to values. Currently, these options are supported:

  • skip_sgml - If true, transliterate SGML tags as if they were ordinary words (<b>iti</b><ब्>इति</ब्>). Defaults to false.
  • syncope - If true, use Hindi-style transliteration (ajayअजय). In linguistics, this behavior is known as schwa syncope. Defaults to false.

Adding new schemes

Adding a new scheme is simple:

Sanscript.addBrahmicScheme(schemeName, schemeData);
Sanscript.addRomanScheme(schemeName, schemeData);

For help in creating schemeData, see the comments on the addBrahmicScheme and addRomanScheme functions.

Something went wrong with that request. Please try again.