All projects

Published, active · Author

Sinhala Phonetic Orthography

A sourced rule set for Sinhala word construction and phonetic romanization.

කගචජටදනපකාගාචාජාටාදානාපාකිගිචිජිටිදිනිපිකීගීචීජීටීදීනීපීකුගුචුජුටුදුනුපුකෙගෙචෙජෙටෙදෙනෙපෙ

Overview

A research study of how Sinhala words are actually built: which letter combinations are valid, which are not, and why, with sources for every rule.

It produced a phonetic romanization that provably never breaks those rules, plus reference code in Python and TypeScript. The rules power Akshara's Smart Phonetic mode.

Highlights

  • 7 topic studies consolidated into 88 rules (hard, soft and style)
  • Validity data for all 923 Sinhala letter forms
  • Romanization verified with 0 rule violations over 3.98 million inputs
  • Frequency-based disambiguator
  • Bilingual site (English and Sinhala) with a playground and letter explorer