NUIG-Panlingua-KMI Hindi-Marathi MT Systems for Similar Language Translation Task @ WMT 2020
Ojha, Atul Kr. Rani, Priya Bansal, Akanksha Chakravarthi, Bharathi Raja Kumar, Ritesh McCrae, John P.
Ojha, Atul Kr.
Rani, Priya
Bansal, Akanksha
Chakravarthi, Bharathi Raja
Kumar, Ritesh
McCrae, John P.
Loading...
Publication Date
2020-11-19
Type
Conference Paper
Downloads
Citation
Ojha, Atul Kr., Rani, Priya, Bansal, Akanksha, Chakravarthi, Bharathi Raja, Kumar, Ritesh, & McCrae, John P. (2020). NUIG-Panlingua-KMI Hindi-Marathi MT Systems for Similar Language Translation Task @ WMT 2020. Paper presented at the 5th Conference on Machine Translation (WMT20) Online, 19-20 November
Abstract
NUIG-Panlingua-KMI submission to WMT 2020 seeks to push the state-of-the-art in the Similar language translation task for the Hindi ↔ Marathi language pair. As part of these efforts, we conducted a series of experiments to address the challenges for translation between similar languages. Among the 4 MT systems prepared for this task, 1 PBSMT systems were prepared for Hindi ↔ Marathi each and 1 NMT systems were developed for Hindi ↔ Marathi using Byte Pair Encoding (BPE) of subwords. The results show that different architectures in NMT could be an effective method for developing MT systems for closely related languages. Our Hindi-Marathi NMT system was ranked 8th among the 14 teams that participated and our Marathi-Hindi NMT system was ranked 8th among the 11 teams participated for the task.
Publisher
Association for Computational Linguistics
Publisher DOI
Rights
Attribution-NonCommercial-NoDerivs 3.0 Ireland