The Seventh ISCA Tutorial and Research Workshop on Speech Synthesis

Kyoto, Japan
September 22-24, 2010

Automatic Prominence Annotation of a German Speech Synthesis Corpus: Towards Prominence-Based Prosody Generation for Unit Selection Synthesis

Andreas Windmann (1), Petra Wagner (1), Fabio Tamburini (2), Denis Arnold (3), Catharine Oertel (1)

(1) Faculty of Linguistics and Literature, Bielefeld University, Germany
(2) Department of Linguistic and Oriental Sdtudies, University of Bologna, Italy
(3) Language and Speech Communication Division, University of Bonn, Germany

This paper describes work directed towarde the development of a syllable prominence-based prosody generation funcionality for a German unit selection speech synthesis system. A general con cept for syllable prominence-based prosody generation in unit selection synthesis is proposed. As a first step towards its implementation, an automated syllable prominence annotation procedure based on acoustic analyses has been performed on the BODD speech corpus. The prominence labeling has been evaluated against an existing annotation of lexical stress levels and manual prominence labeling on a subset of the corpus. We discuss methods and fresults and give an outlook on further implementation steps.

Full Paper

Bibliographic reference.  Windmann, Andreas / Wagner, Petra / Tamburini, Fabio / Arnold, Denis / Oertel, Catharine (2010): "Automatic prominence annotation of a German speech synthesis corpus: towards prominence-based prosody generation for unit selection synthesis", In SSW7-2010, 377-382.