Review of Machine Translation Systems for Verse (Poetic) Translation

Artificial Intelligence (AI) encompasses the field of Natural Language Processing (NLP), which aims to enable machines to perform human-like tasks. This research focuses on Machine Translation Systems that convert poetic texts from Hindi to English. NLP is central to these systems, with key components and stages that guide the translation process. Different types of language translators are discussed, including knowledge-based, corpus-based, and Neural Machine Translation (NMT) systems. NMT, using artificial neural networks, predicts word sequences to improve translation accuracy. The Bilingual Evaluation Understudy (BLEU) score assesses translation quality, which applies a formula to analyze translated text. After reviewing various translators., the study highlights CDEC, a widely used Statistical Machine Translation (SMT) tool for NLP. CDEC is efficient and flexible, offering decoding, alignment, and model training tools. Despite advancements in machine translation, translating poetry, especially from Hindi to English, presents challenges due to the aesthetic and cultural elements embedded in the original text. The review concludes by emphasizing the need to improve existing systems to address these complexities in poetic translations better, such as those of the Ramayana.