Developing a statistical Turkish sign language translation system for primary school students

Abstract

Nowadays, as the access to information in the field of education increases, new technologies are developing for primary school children. However, deaf and dumb children still have limited access to the information especially in their school lives. One of the most important reasons for this problem is the lack of studies in the Turkish Sign Language domain. In this study, for the first time, translation from Turkish to Turkish Sign Language has been performed with statistical machine translation approach. The data required for translation were taken from the textbooks of primary school children and data processing was performed by using various algorithms. The system has been used with Moses Decoder and the results have been tested with different evaluation metrics. Because there is no other SMT based study for Turkish Sign Language in the literature, the results obtained from this study can not be compared. Nevertheless, it is seen that the scores obtained from results are motivating for new studies.