Transliteration is a method of mapping from one system of writing to another based on phonetic similarity.
The model transcripts English language text to Hindi and was built with Tensorflow using GRU unit in encoder and decoder. It is trained on 32000 data points and the data is in uploaded to the repo. The model has a BLEU score of 0.75. BLUE score(range 0-1) measures the similarity of machine translated text to a set of high quality reference translations.