Linguistic and mixed excitation improvements on a HMM-based speech synthesis for Castilian Spanish

Xavier Gonzalvo, Joan-Claudi Socoró, Ignasi Iriondo, Carlos Monzo, Elisa Martínez Marroquín

Research output: A Conference proceeding or a Chapter in BookConference contribution

13 Citations (Scopus)

Abstract

Hidden Markov Models based text-to-speech (HMM-TTS) synthesis is one of the techniques for generating speech from trained statistical models where spectrum and prosody of basic speech units are modelled altogether. This paper presents the advances in our Spanish HMM-TTS and a perceptual test is conducted to compare it with an extended PSOLA-based concatenative (E-PSOLA) system. The improvements have been performed on phonetic information and contextual factors according to the Castilian Spanish language and speech generation using a mixed excitation (ME) technique. The results show the preference of the new HMM-TTS system in front of the previous system and a better MOS in comparison with a real E-PSOLA in terms of acceptability, intelligibility and stability.
Original languageEnglish
Title of host publicationSSW6 - 2007
Subtitle of host publicationProceedings of the Sixth ISCA Workshop on Speech Synthesis
PublisherInternational Speech Communication Association (ISCA)
Pages362-367
Number of pages6
Publication statusPublished - 2007
Externally publishedYes

Fingerprint

Dive into the research topics of 'Linguistic and mixed excitation improvements on a HMM-based speech synthesis for Castilian Spanish'. Together they form a unique fingerprint.

Cite this