WORD BASED STATISTICAL MACHINE TRANSLATION FROM ENGLISH TEXT TO INDIAN SIGN LANGUAGE

The objective of this work is to design a translation machine which can translate English text to Indian Sign Language glosses. A number of rule based approaches are identified in this regard, but our approach is based on statistical machine translation for ISL by using a corpus. The corpus is prepared by collecting glosses and sentences used in Indian Railways for announcement and conversation in public assistance counters. We have experimented this translation with the word based translation model. The methodologies are implemented by the tool MOSES.