Applications of Ergodic Hidden Markov Models in speech synthesis are presented. EHMM using autoregressive gaussian continuous densities as state observation densities are defined. The ability of such model to represent phonotactical constraints of the language is investigated through the analysis of the transition matrix structure and with some ergodic speech synthesis experiments. Applications of EHMM models in speech synthesis units segmentation and phone-like models statistical units representation are presented.
Ergodic Hidden Markov models for speech synthesis / P., Pierucci; Falaschi, Alessandro. - STAMPA. - (1990), pp. 1147-1150. (Intervento presentato al convegno Proceedings of Eusipco-90, Fifth European Signal Processing Conference tenutosi a Barcelona nel settembre 1990).
Ergodic Hidden Markov models for speech synthesis
FALASCHI, Alessandro
1990
Abstract
Applications of Ergodic Hidden Markov Models in speech synthesis are presented. EHMM using autoregressive gaussian continuous densities as state observation densities are defined. The ability of such model to represent phonotactical constraints of the language is investigated through the analysis of the transition matrix structure and with some ergodic speech synthesis experiments. Applications of EHMM models in speech synthesis units segmentation and phone-like models statistical units representation are presented.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
