Sixth ISCA Workshop on Speech Synthesis Bonn, Germany August 22-24, 2007

Utilization of an HMM-Based Feature Generation Module in 5 ms Segment Concatenative Speech Synthesis

Toshio Hirai (1), Junichi Yamagishi (2), Seiichi Tenpaku (2)

(1) Arcadia, Inc., Osaka, Japan
(2) The Centre for Speech Technology Research, University of Edinburgh, UK

If a concatenative speech synthesis system uses more short speech segments, it increases the potential to generate natural speech because the concatenation variation becomes greater. Recently, a synthesis approach was proposed in which very short (5 ms) segments are used. In this paper, an implementation of an HMM-based feature generation module into a very short segment concatenative synthesis system that has the advantage of modularity and a synthesis experiment are described.

Full Paper   Presentation (ppt)   Poster (pdf)

Bibliographic reference.  Hirai, Toshio / Yamagishi, Junichi / Tenpaku, Seiichi (2007): "Utilization of an HMM-based feature generation module in 5 ms segment concatenative speech synthesis", In SSW6-2007, 81-84.