Researchers have developed the first machine translation systems for the English and Pnar language pair, an Austroasiatic language spoken by approximately 0.4 million people. Using articles from the Wyrta newspaper, a parallel corpus of 10,234 sentences was created to train phrase-based statistical machine translation models. The best performing system achieved a BLEU score of 14.97 for Pnar to English and 11.16 for English to Pnar, establishing a quantitative benchmark for this low-resource language pair. AI
IMPACT Establishes a benchmark for machine translation of low-resource languages, potentially guiding future NLP development for similar languages.
RANK_REASON Academic paper detailing a new machine translation system for a low-resource language pair. [lever_c_demoted from research: ic=1 ai=1.0]
- BLEU
- English
- GIZA++
- KenLM
- minimum error rate training (MERT)
- Moses
- Pnar
- Statistical Machine Translation
- Thoudam Doren Singh
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →