A step towards Torwali machine translation: an analysis of morphosyntactic challenges in a low-resource language

摘要

Torwali is an endangered language spoken in the north of Pakistan. It is a computationally challenging language because of its RTL Perso-Arabic script, non-concatenative nature and distinct words alterations. This paper discusses issues and challenges regarding grammatical structure, divergence in terms of lexicon as well as morphological makeup for the machine translation of a less studied language. It includes creation of NLP tools such as parts of speech (POS) tagger and morphological analyser with HFST which is based on the idea of building lexicon and morphological rules using finite state devices. This work, on which this paper is based, will be a source of Torwali finite state morphology and its future computational growth as electronic dictionaries are usually equipped with morphological analyser and it will also be helpful for developing language pairs.