Google India has launched a toolkit for Indian language translation to help spur the addition of Indic content online. This language editing application is a follow up to Google Translate, which automatically translates websites or documents to Hindi, and its Indic Transliteration initiative, which lets users easily type in Hindi, Kannada, Bengali, Gujarati, Tamil, Telugu, Marathi, Nepali and Malayalam. There are currently 5 Indic languages available – Hindi, Kannada, Tamil, Telugu and Malayalam. Note that it only offers translation from English sources.

Will It Be Used? Implications

Google Translate has been around for a few years and is fine for a sketchy literal translation or a hurried read; the transliteration application services primarily bilingual users. The Translator Toolkit will enable fine tuning of Google Translate, giving it a larger, more comprehensive library that will offer complete interpretations to all the words and sentences that can be translated to Indic languages.

[ Nikhil adds: If it doesn't do this already, Google can determine the context of queries in Indic languages better, based on translated versions of pages; we don't know if they already do this or not.  At present, Google does end up catering primarily to bilingual users with its Indic language search. Question is - will Indians actually spend time on improving Google Translate? I remember Jimmy Wales, a couple of years ago, talking about the abysmal state of Indian language updates to Wikipedia. ]

The Crowdsourcing Model

Google has chosen the “crowdsourcing” route to Indian language translation, much like they did with Google MapMaker last year, for improving Google Maps. Even as users edit translations of content using the toolkit, their changes will be stored, shared and added via a feedback mechanism to Google Translate’s library. A video of its usage is on YouTube.

As a standalone system, the toolkit will help users add quality content – they can first use Google Translate to automatically translate most of the English content to their language, then add, correct or edit it using various tools such as a dictionary, multilingual glossary and placeholders. Another tool available is translation memory, which is a database of previously recorded translations by users.

Related

- Google Introduces Transliteration Bookmarklets For 5 Indian Languages
-All stories on Indic languages.