Open machine translation for low resource South American languages (AmericasNLP 2021 shared task contribution)
- Publication type
- C1
- Publication status
- Published
- Authors
- Parida, S., Panda, S., Dash, A., Villatoro-Tello, E., Doğruöz, A.S., Ortega-Mendoza, R., Hernández, A., Sharma, Y., & Motlicek, P.
- Editor
- Manuel Mager, Arturo Oncevay, Annette Rios, Ivan Vladimir Meza Ruiz, Graham Neubig and Katharina Kann
- Series
- Proceedings of the First Workshop on Natural Language Processing for Indigenous Languages of the Americas
- Pagination
- 218-223
- Publisher
- Association for Computational Linguistics (ACL)
- Conference
- First Workshop on Natural Language Processing for Indigenous Languages of the Americas (NAACL-HLT 2021) (Online (Mexico City, Mexico))
- Download
- (.pdf)
- View in Biblio
- (externe link)
Abstract
This paper describes the team (“Tamalli”)’s submission to AmericasNLP2021 shared task on Open Machine Translation for low resource South American languages. Our goal was to evaluate different Machine Translation (MT) techniques, statistical and neural-based, under several configuration settings. We obtained the second-best results for the language pairs “Spanish-Bribri”, “Spanish-Asháninka”, and “Spanish-Rarámuri” in the category “Development set not used for training”. Our performed experiments will serve as a point of reference for researchers working on MT with low-resource languages.