Blog
Blog
Long-form articles on transliteration engineering and research.
- 2026-09-05 The Arabic frontier, closed: every rung measured, two verdicts negative, one 95 MB browser model == Introduction Two weeks ago the phonological layer shipped with a disclosed gap: the Arabic client model scored 8.26 on the full benchmark. This week
- 2026-08-24 The phonological layer: ten models, three wrong turns, and one correction == Introduction Interscript has always been about deterministic transliteration: 289 authority-backed systems, one map format, byte-identical output from
- 2022-04-04 Learning transliteration from graphs == Introduction === General Evaluating existing transliteration technologies for Farsi, we found the
- 2021-10-03 Extending Rababa for Hebrew Diacriticization == Introduction === General Rababa is an automated diacriticization
- 2021-08-03 Diacritization in Arabic with Deep Learning == Introduction === General Diacritization is the art of completing Arabic scripts with the correct
- 2021-06-26 WebAssembly and advanced regular expressions with Opal NOTE: This is also published as a guest post at the Opal website. Thank you and a great hug to the Opal team!