Grapheme-to-Phoneme Conversion for Amharic Text-to-Speech System

Developing correct Grapheme-to-Phoneme (GTP) conversion method is a central problem in text-tospeech synthesis. Particularly, deriving phonological features which are not shown in orthography is challenging. In the Amharic language, geminates and epenthetic vowels are very crucial for proper pronunciation but neither is shown in orthography. This paper describes an architecture, a preprocessing morphological analyzer integrated into an Amharic Text to Speech (AmhTTS) System, to convert Amharic Unicode text into phonemic specification of pronunciation. The study mainly focused on disambiguating gemination and vowel epenthesis which are the significant problems in developing Amharic TTS system. The evaluation test on 666 words shows that the analyzer assigns geminates correctly (100%). Our approach is suitable for languages like Amharic with rich morphology and can be customized to other languages.