A Study on Approaches to Neural Machine Translation
Journal of Logistics Informatics and Service Science · 2022
Machine translation is a computer-oriented translation procedure that takes a set of words belonging to a specific human readable language as input and returns another set of words in the desired human readable language as output.The advent of machine translation has benefited sociologists and linguists across the globe.A machine translation model can be, Rule-based, Statistical based, or Neural Network based.Due to the multiple limitations associated with rule-based and statistical-based machine translation models, Neural Machine Translation (NMT) model came into the picture.Machine translation models based on neural networks showed promising results for different human-readable languages, with a rich vocabulary trained with a large corpus.This paper aims to present a literature study on different approaches for different operations in Neural Machine Translation, such as word embedding, attention mechanisms, etc.Also, we emphasized our survey on Neural Machine Translation systems dedicated to Indian languages.