An Enhanced English to Marathi Translator using sequence-to-sequence Transformer

Machine translation is a key application in study of natural language processing that enables the translation of text or speech between different languages automatically. In this research, we provide a brand-new sequence-to-sequence transformer model for translating between English and Marathi. Our model can produce translations that are accurate and fluid because it was trained on a big dataset of parallel English-Marathi utterances. We used a test set to analyse the performance of our model, and the result was a 41.99 BLEU score. Our findings show that even on low-resource datasets for English to Marathi language translation, our model performs well.