The parser from an Arabic text-to-speech system
Allan M. Ramsay, Hanady Mansour · 2004
The work described here is part of an attempt to provide a text-to-speech (TTS) system for Modern Standard Arabic (MSA). The key problem with this task is that written MSA omits a great deal of information about short vowels, and a certain amount of other phonetically relevant information, which is clearly essential for any speech synthesiser. We have argued elsewhere that in order to recover this information you need to carry out a detailed linguistic analysis at a variety of levels. We have described elsewhere how we integrate morphological, syntactic and semantic processing. The current paper provides a detailed description of our approach to syntactic analysis of MSA. The presentation here assumes that morphological processing is interwoven with syntactic analysis, but the details of our approach to morphology are omitted: the interested reader is invited to see (Ramsay and Mansur2001). 1