Translating the whole Ur III corpus

This project aims to build a full translation pipeline with the integration of NER (Named Entity Recognition) and POS (Part of Speech) tagging of Sumerian language and Post-processing of the translated English text to produce best and accurate results. The model should be built either using rule based or Neural based approach. The final pipeline including the Machine Translation System (Different Project) should be able to tag and translate the whole 1.5M Sumerian text perfectly.