A Bilingual Machine Transliteration System for Sanskrit-English Using Rule-Based Approach

Machine Transliteration is a big challenging area in an increasingly multilingual ecosphere due to its dangerous position in various downstream natural language processing application systems. When a word is transliterated, it is shifted from one script to another. In contrast to a translation, which clarifies the meaning of a word written in a different language, a transliteration only conveys the word's pronunciation by utilizing a familiar alphabet. This paper proposes a technique for creating a bilingual automated tool to type Sanskrit using English orthography/alphabets and converting Sanskrit text into the script of English which helps in reading the Sanskrit text for those who are not aware about the orthography of Sanskrit language. The system receives input via the QWERTY keyboard, which produces the equivalent Sanskrit text and inversely user can give Sanskrit text as input to get equivalent text using the script of English language. The goal is to create an easy-to-use and robust automated solution that allows end-users to effortlessly type Sanskrit shlokas or sentences using an English keyboard. The suggested approach is unique in that it is based on the language's Unicode and works for the low-resource ancient language Sanskrit. The primary applications of the designed tool are to help user to read Sanskrit text using the script/orthography of English, to create parallel corpora for the Sanskrit language translation process, to create e-versions of manuscripts written in Sanskrit language as the majority of ancient knowledge is not available on the internet, and to make it easy to learn Sanskrit language by any individual or students at the school level. This tool encourages humans to use their original language.