And now Mentamizh

The word processor for the Tamil language Mentamizh, is the result of 10 years of intensive research in computational linguistics of Professor N. Deivasundaram, Tamil professor and computational linguist, who retired from the University of Madras, in 2010.

“Along with my team that comprises Tamil language experts, computational linguists, computer engineers, we have successfully developed the first ever word processor for Tamil language called Mendamizh,” Prof. Deivasundaram says. So far we have had Tamil keyboards and fonts, with which one could type in Tamil.

With this software, writing, editing and publishing in Tamil language has been made easier. Using Mentamizh one can type in Tamil (and English) using the rich text editor and with all features of modern word processor. The most important feature of this software is the spell checker and sandhi checker, a tool that can check the Tamil document for any phonemic error. The integrated Tamil dictionary in Mentamizh has 42,000 modern Tamil words with English equivalences. “Another unique aspect that we have built-in is mayangoli agaradhi (suspicious pair dictionary),” says Prof. Deivasundaram. This enables the user to get some clarification with regard to certain words with suspicious phonemes.

Apart from supporting multiple file formats and 11 different types of Tamil keyboards, users of Mentamizh need not worry about the non Unicode documents. All old fonts (prior to Unicode) can be converted to Unicode in single stroke. Tools used for publishing such as citation, bibiliography and table of contents are available. Words in Tamil document could be indexed with the page numbers.

Bringing Tamil morphology and phonology into computing has been a challenging task, but the outcome of our team efforts has been excellent, says Prof. Deivasundaram. His research is now focused on machine translation of Tamil, wherein text to speech and speech to text will become possible. This will benefit the hearing and speech impaired and visually challenged students immensely.

Tamil has to be developed into an e-language. Only then will this ancient language grow and evolve. Otherwise, it will become an endangered language, he says. Among the Indian languages, only Hindi and Sanskrit have been well developed as e-languages.