Transliteration Model for Egyptian Words

Abstract

In this paper, we describe token-based transliteration models for Egyptian words. We explain how we created them using an automatic alignment method we devised based on the Needleman-Wunsch sequence alignment algorithm. We use two sources where encoded Egyptian hieroglyphs and their transliteration pairs are available. Ancient Egyptian Sentences (AES) includes a collection of texts where c. 254,000 Egyptian words encoded using Manual de Codage (MdC) have been aligned with their transliteration counterparts. The second source is the Ramses Transliteration Corpus (RTC), with almost 500,000 MdC encoded words. The RTC consists of encoded hieroglyphic sentences, each on its line, and respective transliteration lines in another file. However, unlike the AES, there is no ready alignment of the MdC and its transliteration on the word level. In order to find word-transliteration pairs, we align the sentences of encoded words with the respective transliterations. The alignment task is made more difficult because many of the texts contain damaged parts and editorial additions.