The Hindu : National : Computers to help convert Santali books

Online edition of India's National Newspaper
Friday, Apr 27, 2007
ePaper


<![if lt IE 4]> <script src="http://servedbyadbutler.com/adserve/;ID=150167;size=468x60;setID=96691;type=js" type="text/javascript"> </script> <noscript> <a href="http://servedbyadbutler.com/go2/;ID=150167;size=468x60;setID=96691"> <img src="http://servedbyadbutler.com/adserve.ibs/;ID=150167;size=468x60;setID=96691;type=img" border="0" height="60" width="468" alt=""></a> </noscript><![endif]>

National

National

Computers to help convert Santali books

Antara Das

Books in Bengali or Devanagari script can be converted to Ol Chiki script

Kolkata: Computer scientists, applying the latest in technology, may soon be able to turn the tide in favour of Santali, a language spoken by at least 10 million people mainly in the eastern and north-eastern parts of India.

If the proposed project — at the Computer Science and Engineering (CSE) Department at Jadavpur University — reaches fruition, it will be possible to convert Santali books, written in Bengali or Devanagari script, to the native Ol Chiki script with minimum human intervention.

"To begin with, the Optical Character Recognition (OCR) will convert the Bengali or Devanagari (Hindi) characters to an electronic version while the errors committed in the process will be rectified through an error correction interface embedded in the system," said Anirban Raychaudhuri, research scientist, at Centre of Microprocessor Application Education Training and Research, CSE.

A transliterator module will then convert it into the Ol Chiki script, further errors being corrected through another interface and inputs by a Santali language expert, he added.

Technological development, which has so far been limited to the major Indic languages.

While the required OCR software has been designed as part of a Department of Science and Technology project, a multi-script editor for Santali that includes the Ol Chiki script has also been developed as part of a World Bank-funded project in a collaboration between CSE and the University's Comparative Literature Department.

Along with cooperation by the e-group Wesanthals, the projects received the encouragement of the former department head, Dr. Mita Nasipuri.

The Centre is also planning a 1,000 word strong multilingual glossary with speech data support in Santali for exact pronunciation, as well as a Santali language training module in a compact disc format.

Though spoken by a large number of people, Santali has a low literacy rate of 25 per cent and is neither a language of knowledge or political and economic power.

For the tribal language speaker, the use of the mother tongue often becomes a hindrance to personal advancement and upward mobility, leading to its large-scale abandonment.

Printer friendly page
Send this article to Friends by E-Mail

National