We use cookies on our site to analyze traffic, enhance your experience, and provide you with tailored content.
For more information visit our privacy policy.Statistical Machine TranslationStatistical Machine Translation Evaluation • August 4th, 2022
Contract Type FiledAugust 4th, 2022This paper contributes the first evalu- ation of the quality of statistical ma- chine translation (SMT) between Myan- mar (Burmese) and Dawei (Tavoyan). We also developed a Myanmar-Dawei parallel corpus (around 9K sentences) based on the Myanmar language of ASEAN MT corpus. The 10 folds cross-validation experiments were carried out using three different sta- tistical machine translation approaches: phrase-based, hierarchical phrase-based, and the operation sequence model (OSM). In addition, two types of segmentation were studied: word and syllable segmen- tation. The results show that all three statistical machine translation approaches give comparable BLEU and RIBES scores for both Myanmar to Dawei and Dawei to Myanmar machine translations. OSM approach achieved the highest BLEU and RIBES scores among three SMT ap- proaches for both word and syllable seg- mentation.