Eighth ISCA Workshop on Speech Synthesis

Barcelona, Catalonia, Spain
August 31-September 2, 2013

Is Unit Selection Aware of Audible Artifacts?

Jindřich Matoušek, Daniel Tihelka, Milan Legát

University of West Bohemia, Faculty of Applied Sciences, Department of Cybernetics, Czech Republic

This paper presents a new analytic method that can be used for analysing perceptual relevance of unit selection costs and/or their sub-components as well as for tuning of the unit selection weights. The proposed method is leveraged to investigate the behaviour of a unit selection based system. The outcome is applied in a simple experiment with the aim to improve speech output quality of the system by setting limits on the costs and their sub-components during the search for optimal sequences of units. The experiments reveal that a large number (36.17 %) of artifacts annotated by listeners are not reflected by the values of the costs and their sub-componets as currently implemented and tuned in the evaluated system. Index Terms: speech synthesis, unit selection, concatenation cost, target cost, audible artifacts

Full Paper

Bibliographic reference.  Matoušek, Jindřich / Tihelka, Daniel / Legát, Milan (2013): "Is unit selection aware of audible artifacts?", In SSW8, 267-271.