A Large-scale Evaluation of Neural Machine Transliteration for Indic Languages
Anoop Kunchukuttan, Siddharth Jain, Rahul Kejriwal · 2021
We take up the task of largescale evaluation of neural machine transliteration between English and Indian languages, with a focus on multilin gual transliteration to utilize orthographic sim ilarity between Indian languages.We create a corpus of 600K word pairs mined from parallel translation corpora and monolingual corpora, which is the largest transliteration corpora for Indian languages mined from public sources.We perform a detailed analysis of multilingual transliteration and propose an improved mul tilingual training pipeline for Indic languages.We analyse various factors affecting transliter ation quality like language family, translitera tion direction and word origin.