Sinhala Transliteration: A Comparative Analysis Between Rule-based and Seq2Seq Approaches

TL;DR

This paper compares rule-based and Transformer-based sequence-to-sequence methods for Romanized Sinhala transliteration, demonstrating that the Transformer approach captures complex patterns more effectively.

Contribution

It introduces a Transformer-based neural transliteration model for Sinhala, outperforming traditional rule-based methods in capturing script patterns.

Findings

Transformer method captures more complex patterns

Transformer outperforms rule-based approach

Code available on GitHub

Abstract

Due to reasons of convenience and lack of tech literacy, transliteration (i.e., Romanizing native scripts instead of using localization tools) is eminently prevalent in the context of low-resource languages such as Sinhala, which have their own writing script. In this study, our focus is on Romanized Sinhala transliteration. We propose two methods to address this problem: Our baseline is a rule-based method, which is then compared against our second method where we approach the transliteration problem as a sequence-to-sequence task akin to the established Neural Machine Translation (NMT) task. For the latter, we propose a Transformer-based Encode-Decoder solution. We witnessed that the Transformer-based method could grab many ad-hoc patterns within the Romanized scripts compared to the rule-based method. The code base associated with this paper is available on GitHub - https://github.com/kasunw22/Sinhala-Transliterator/