Transliteration at its best, Google does it again.

I just stumbled upon this page on Google Beta labs that offers indic transliteration for several languages. The languages supported by this tool include Urdu, Bengali, Gujarati, Hindi, Kannada, Malayalam, Marathi, Nepali, Punjabi, Tamil and Telugu.

All of us have grown up using Internet in only one language. Now is the time, things are changing and a bunch of language options are available not only in operating systems and software, but also on popular web systems. But it took a lot of time to reach this point and all of us (well most of us) have already learnt to type our own languages using English character set.

Now, Google’s not only providing translation solutions, but transliteration is also gaining popularity. I have tried the solution out on Urdu language and the results are absolutely amazing. Its not only converting most of the words correctly, its also capable of handling the mystery of using different versions of ( ح – ہ ) or being sure when to use ت and when to use ط . For example, it correctly transliterates “bht” to “بھت” and “tht” to “تحت” . It also adds letters that are usually silent when typing in English. For example “Talat” is correctly transliterated to “طلعت” converting T correctly to ط , not ت and introducing ع while it was not obvious in English character version.

Following are a few results that really amazed me:

Chhachhoondar >>> چھچھوندر

Pakistan ka matlab kya la ilaha illalla >>> پاکستان کا مطلب کیا لا الہٰ الاللہ

Ap sab is pabandi say mustansna hain >>> اپ سب اس پابندی سے مستثنیٰ ہیں

Mustansar Husain Tarar >>> مستنصر حسین تارڑ

Chacha nay chachi ko chandni chowk main chandi ki chamchi say chatni chataai >>> چچا نے کچی کو چاندنی چوک میں چاندی کی چمچی سے چٹنی چٹائی

(This one was more of a riddle, used as a tounge twister usually)

Well, its an automated system and there are still some issues. For example it could not transliterate my name, is’nt that pretty simple?. 🙂 Better luck next time Google!