Patiala: Indians able to read only Devnagri can now study Persian, Urdu, Punjabi or Sindhi literature from Pakistan, and vice-versa.
Punjabi University has developed the first Perso-Arabic to Indic script transliteration system, which has the power to break the script barrier between the two countries and their common languages. Developed by Gurpreet Singh Lehal and his team, the software can convert with high accuracy any text from Urdu to Devnagri; Sindhi Arabic to Sindhi Devnagri; and Shahmukhi to Gurmukhi, and the other way round. Five international agencies based in Singapore, Australia, Sweden and Canada funded the project art different stages over 10 years.
The software is available for downloading at sangam.learnpunjabi.org. “It will give people an opportunity to enlighten themselves with literature and other readable information in different languages. The development cost was Rs 50 lakh,” said Lehal. Defense Research and Development Organisation, which makes technology for the Indian defence system and is keen to decode messages in Urdu and other languages, contributed Rs 10 lakh.
“Copy and paste any Urdu text in the software, you will get it transliterated in Devnagri script. So even without recognising letters of the Urdu alphabet, you can read the original article,” said Lehal. University’s dean of academic affairs AS Chawla launched this software during the third regional international conference on natural language processing, where Dipti Mishra Sharma of the Indian Institute of Information Technology (IIIT) talked about the importance of language as a source of knowledge about culture. “To save dying Indian-subcontinent languages, we need technology for natural language processing and building computational language resources,” she said.
Participating professor Neeraj Sharma said the university had targeted six North Indian languages - Hindi, Punjabi, Urdu, Dogri, Sindhi and Kashmiri - for developing natural language processing software.