Sixth European Conference on Speech Communication and Technology
Text preprocessing plays an important role in a text-tospeech (TTS) synthesis system. The correct detection and interpretation of input strings influence the overall system accuracy and contribute to the conversion of an unrestricted text into synthetic speech. This paper describes the design philosophy of a preprocessing module for a TTS system in Romanian language. The preprocessor is implemented using the standard flex/bison lexer and parser generators. The paper discusses the text preprocessing task and the major difficulties connected with Romanian language, proposes a set of definitions and rules, gives some implementation details and concludes with a few considerations about the TTS system and performances of the preprocessing module.
Full Paper (PDF) Gnu-Zipped Postscript
Bibliographic reference. Burileanu, Dragos / Dan, Claudius / Sima, Mihai / Burileanu, Corneliu (1999): "A parser-based text preprocessor for romanian language TTS synthesis", In EUROSPEECH'99, 2063-2066.