5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

On the Use of Automatically Generated Discourse-Level Information in a Concept-to-Speech Synthesis System

Janet Hitzeman (1), Alan W. Black (1), Paul Taylor (1), Chris Mellish (2), Jon Oberlander (3)

(1) Centre for Speech Technology Research, University of Edinburgh, UK
(2) Department of Artificial Intelligence, University of Edinburgh, UK
(3) Human Communication Research Centre, University of Edinburgh, UK

This paper describes the latest version of the SOLE concept-to-speech system, which uses linguistic information provided by a natural language generation system to improve the prosody of synthetic speech. We discuss the types of linguistic information that prove most useful and the implications for text-to-speech systems.

Full Paper

Bibliographic reference.  Hitzeman, Janet / Black, Alan W. / Taylor, Paul / Mellish, Chris / Oberlander, Jon (1998): "On the use of automatically generated discourse-level information in a concept-to-speech synthesis system", In ICSLP-1998, paper 0591.