Parallel Corpora Preparation for English-Amharic Machine Translation

Authors

Contributors

Bahir Dar Institute of Technology (BiT); Bahir Dar University (BDU); Statistical Machine Translation and Speech Modelization and Text (SMarT); Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD); Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA); Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA); Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS) + 5

Published in

IWANN 2021 - International Work on Artificial Neural Networks, Conference Spr... >

Publisher

HAL CCSD

Abstract

International audience ; In this paper, we describe the development of an English-Amharic parallel corpus and Machine Translation (MT) experiments conducted on it. Two different tests have been achieved. Statistical Machine Translation (SMT) and Neural Machine Translation (NMT) experiments. The performance using the bilingual evaluation understudy metric (BLEU) shows 26.47 and 32.44 respectively for SMT and NMT. The corpus was collected from the Internet using automatic and semi automatic techniques. The harvested corpus concerns domains coming from Religion, Law, and News. Finally, the corpus, we built is composed of 225,304 parallel sentences, it will be shared for free with the community. In our knowledge, this is the biggest parallel corpus so far concerning the Amharic language.