IN THE VERBMOBIL PROJECT Simon Kingy Thomas Portele Florian H ofer Institut f ur Kommunikationsforschung und Phonetik (IKP), Universit at Bonn Poppelsdorfer Allee 47, D-53115 Bonn, Germany http://www.ikp.uni-bonn.de ynow at the Centre for Speech Technology Research, University of Edinburgh, 80, South Bridge, Edinburgh EH1 1HN, GB http://www.cstr.ed.ac.uk email: Simon.King@ed.ac.uk ABSTRACT We describe a concatenative speech synthesiser for British English which uses the HADIFIX [8] inventory structure originally developed for German by Portele. An inventory of non-uniform units was investigated with the aimof improving segmental quality compared to diphones. A combination of soft (diphone) and hard concatenation was used, which allowed a dramatic reduction in inventory size. We also present a unit selection algorithm which selects an optimum sequence of units from this inventory for a given phoneme sequence. The work described is part of the concept-to-speech synthesiser for the language and speech project Verbmobil [12] which is funded by the German Ministry of Science (BMBF).
[1]
Alan W. Black,et al.
CHATR: a generic speech synthesis system
,
1994,
COLING.
[2]
Stephen Isard,et al.
Optimal coupling of diphones
,
1994,
SSW.
[3]
Paul Taylor,et al.
Assigning intonation elements and prosodic phrasing for English speech synthesis from high level linguistic input
,
1994,
ICSLP.
[4]
Alan W. Black,et al.
Synthesizing conversational intonation from a linguistically rich input
,
1994,
SSW.
[5]
D. Whalen.
Coarticulation is largely planned
,
1990
.
[6]
Wolfgang Wahlster,et al.
Verbmobil: Translation of Face-To-Face Dialogs
,
1993,
MTSUMMIT.
[7]
Julia Hirschberg,et al.
Using discourse context to guide pitch accent decisions in synthetic speech
,
1990,
SSW.
[8]
Wolfgang Hess,et al.
A mixed inventory structure for German concatenative synthesis
,
1994,
SSW.