1st Joint SIG-IL/Microsoft Workshop on Speech and Language Technologies for Iberian Languages

Porto Salvo, Portugal
September 3-4, 2009

A Detailed Analysis and Comparison of Speech Synthesis Paradigms

Luis Coelho (1), Daniela Braga (2), Carmen Garcia-Mateo (3)

(1) ESEIG, Instituto Politecnico do Porto, Porto, Portugal
(2) Microsoft Language Development Center, Microsoft, Portugal
(3) Departamento de Teora de la Seal y Comunicaciones, University of Vigo, Spain

Hidden Markov Model based synthesis has gained a special relevance in international speech conference and in contests like the Blizzard Challenge. This developing technology has proved to be quite promising but naturalness is still an achievement to conquer. In this paper we compare two TTS technologies with a human speaker and provide some of the major observed similarities and differences. Our analysis covers time and frequency domains for several acoustical units in order to demonstrate, in this short space, the capabilities of each technology.

Index Terms: speech synthesis, comparison, HMM synthesis

Full Paper

Bibliographic reference.  Coelho, Luis / Braga, Daniela / Garcia-Mateo, Carmen (2009): "A detailed analysis and comparison of speech synthesis paradigms", In SLTECH-2009, 71-74.