Sixth European Conference on Speech Communication and Technology
In text-to-speech (TTS) synthesis, input text is automatically analyzed. This involves prediction of pronunciation, intonation, and timing at segmental and phrase level. In the design of dialog applications, developers need more control over the text-to-speech conversion. While the automatic analysis is often unsatisfactory, the developer can easily provide hints that improve the synthetic speech. The Sable markup language, which has been proposed as a standard for TTS, includes tags to indicate emphasis, speaking rate, phrase breaks, and other properties. We extend this work as follows. First, we describe a graphical editor (GUI) for Sable. An interesting challenge is to find intuitive mappings between the visual representation of the text and the attached markup properties. Next, we discuss the addition of several new markup commands and the implementation in Festival, the TTS platform we use. Finally, we describe our experiences using the authoring tools in a language training project for profoundly deaf children. The authoring tools are made freely available viahttp://cslu.cse.ogi.edu/tts.
Full Paper (PDF) Gnu-Zipped Postscript
Bibliographic reference. Wouters, Johan / Rundle, Brian / Macon, Michael W. (1999): "Authoring tools for speech synthesis using the sable markup standard", In EUROSPEECH'99, 963-966.