By
Clement Odoje
Department of Linguistics and African Languages
University of Ibadan
And
Solomon O. Akinola
Computer Engineering,
College of Engineering,
Afe Babalola University,
P.M.B. 5454,
Ado Ekiti
Abstract
The challenges of Machine Translation (MT) and in particular Statistical Machine Translation (SMT) have been explored and categorized. But little is known about African languages which are said to be resource scare languages. Hence, this paper explored the challenges of SMT for African languages using English-Yoruba MT as case study. Fagunwa's books and its English translated equivalent versions were used as corpus and Moses was used as the language toolkit. While the challenges were inexhaustible it was found that the challenges of Africa SMT can be categorized into two: technical and sociocultural determinants.