Researchers have developed the first phrase-based Statistical Machine Translation (SMT) model for English-to-Syriac translation, addressing the language's status as endangered and understudied in Natural Language Processing (NLP). The study created a dataset of nearly 40,000 sentence pairs from the English and Syriac Bible, employing custom scripts for data extraction and manual alignment. The best-performing model achieved a BLEU score of 23.54, with human evaluations showing adequate and fluent translations. This work provides a publicly available dataset and a baseline for future NLP research on Syriac. AI
IMPACT Provides a foundational dataset and model for NLP research on endangered languages, potentially enabling new applications.
RANK_REASON Academic paper detailing a new machine translation model and dataset for a low-resource language. [lever_c_demoted from research: ic=1 ai=1.0]
- Assyrian Neo-Aramaic
- Bible
- Bleu
- byte-pair encoding
- East Syriac Dialect
- English
- machine translation
- Madnkhaya script
- Moses framework
- natural language processing
- Syriac
- UNESCO
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →