This paper, discusses the approach adopted in the development of a bidirectional Machine Translation system for Indian languages. The approach makes use of the characteristics of the languages in simplifying the process of translation. The verb- final sentence structure and the case-inflected nature of Indian language sentences have led us to adopt a verb-centered approach. The analysis is carried out at phrase level, with a phrase being demarked by a verb. A frame-based representation scheme is used to map the results of the analyzer. Due to the free word order nature of Indian languages, the word order of the source languages is retained in the target language. The system translates the typical conversation at railway counters between Telugu and Hindi. Keywords: verb-centered approach, case inflections, frames, free word order
Cite as: Subramaniam, N.V., Alwar, N., Mallikarjuna, G., Rao, P.P., Raman, S. (1991) Bidirectional machine translation in indian languages. Proc. 2nd European Conference on Speech Communication and Technology (Eurospeech 1991), 1067-1070, doi: 10.21437/Eurospeech.1991-254
@inproceedings{subramaniam91_eurospeech,
author={N. Venkata Subramaniam and N. Alwar and G. Mallikarjuna and P. Prabhakar Rao and S. Raman},
title={{Bidirectional machine translation in indian languages}},
year=1991,
booktitle={Proc. 2nd European Conference on Speech Communication and Technology (Eurospeech 1991)},
pages={1067--1070},
doi={10.21437/Eurospeech.1991-254}
}