Indian Govt to Build Machine Translation System for 22 Indic Languages

Brindaalakshmi's article was published by Medianaama on November 14, 2013. CIS memorandum of understanding with Goa University is mentioned.

The report states that scientific and technical terminologies are being developed in 69 disciplines of study in 22 languages as part of this project. Apparently, 25,000 to 30,000 lexicons have been identified in each subject to create the MTS.

Speaking at a media event, Prof V Saratchandran Nair, Director of National Translation Mission said that the project is being initiated to help students who have done their schooling in vernacular languages but have to study in English during college.  He also added that only Tamil currently has translated materials in all 69 disciplines, while Oriya has it in 49 disciplines. In contrast, there is no science material available in Santhali, Manipuri and Konkani languages.

National Translation Mission is a Government of India initiative that intends to establish translation as an industry and facilitate higher education by making available translated study material for students. The program also trains translators in different languages. Interestingly, the website states that it will not replicate the efforts of other organisations like C-DAC, TDIL that are also involved in creating indic support for computers and services. However, it seeks to build technology to directly translate a sentence typed in English into an Indian language. When we checked, we noticed that the website also offers dictionaries and other resources for translators.

Other efforts to provide indic support: In September 2013, the Goa university had entered into a three year memorandum of understanding (MoU) with the Centre for Internet and Society (CIS) for building the Konkani Wikipedia. As part of this partnership, Goa University will be uploading the four volumes of Konkani encyclopedia. This could be a good start to building a repository of translated works for science in Konkani.

In July 2013, Technology Development for Indian Languages Programme (TDIL) of DeitY (Department of Electronics and IT) had developed Urdu language fonts and keyboard drivers for Windows and Android.

In June 2013, DeitY had also created a repository of fonts for all 22 constitutionally recognized languages through TDIL. These fonts are available as a CD that can be procured from TDIL or the entire suite can also be downloaded from the TDIL website.