A web based Punjabi to Hindi Statistical Machine Translation System

Abstract

Machine translation (MT) system translates one natural language into another language with the help of computers. MT is a key application in the field of natural language processing. In proposed system, Statistical Machine Translation (SMT) approach has been used for developing Punjabi to Hindi Machine Translation System. In SMT, every sentence in the target language i.e. Hindi is a translation of the source language i.e. Punjabi with some probability and the best translation will be of high probability which the system will attain in the form of sentence. The key activities involved during translation process are pre-processing, translation engine and post processing. Unigram algorithm, N-Gram string matching algorithm etc formed the basis for solving these issues. The accuracy of the system has been evaluated using subjective tests i.e. intelligibility test and accuracy test. A proposed system is found to perform better than the existing system.