The experimental part of Google never ceases to amaze – and their latest handiwork would mean you will be able to explore the linguistic diversity of the Indian sub-continent with Google Translate, where it will be able to support five new experimental alpha languages including Bengali, Gujarati, Kannada, Tamil and Telugu. It is estimated that just India and Bangladesh alone have over 500 million people who speak these five languages. From 2009 onwards, Google has launched a total of 11 alpha languages, which means the total number of languages supported by Google Translate is now an extremely respectable 63.
Since Indic languages does differ from English in many ways, it is definitely a fresh and welcome challenge to developers when it comes to the translation system. Indian languages tend to use the Subject Object Verb (SOV) ordering to form sentences, which is different from English that uses Subject Verb Object (SVO) ordering. Of course, nothing cannot be solved with mathematics (except love, as the old French proverb says that ” Try to reason about love and you will lose your reason”), and translation definitely falls under the mathematical field.
Well, we are sure that future versions of Google Translate will be able to be more accurate as with current mature languages such as Spanish or Chinese, since those two already have more web content that will assist the statistical machine translation approach.