Himanshu Choudhary, Aditya Kumar Pathak, Rajiv Ratn Shah, Ponnurangam Kumaraguru, EMNLP WMT (2018).


A huge amount of valuable resources is avail- able on the web in English, which are of- ten translated into local languages to facilitate knowledge sharing among local people who are not much familiar with English. How- ever, translating such content manually is very tedious, costly, and time-consuming process. To this end, machine translation is an effi- cient approach to translate text without any hu- man involvement. Neural machine translation (NMT) is one of the most recent and effective translation technique amongst all existing ma- chine translation systems. In this paper, we apply NMT for English-Tamil language pair. We propose a novel neural machine translation technique using word-embedding along with Byte-Pair-Encoding (BPE) to develop an ef- ficient translation system that overcomes the OOV (Out Of Vocabulary) problem for lan- guages which do not have much translations available online. We use the BLEU score for evaluating the system performance. Ex- perimental results confirm that our proposed MIDAS translator (8.33 BLEU score) outper- forms Google translator (3.75 BLEU score).