Application to Many Languages:
Multilingual HMM-based Speech Synthesis

Demonstrations

Languages
Spanish
Austrian German
Finnish
Mandarin
Japanese
Speaker 1
Speaker 2
Speaker 3

Languages
Hindi
Romanian
Ibibio
Speaker 1



Acknowledgement

These voices have been built through collaboration and cooperation with many researchers
Spanish voices
R. Barra-Chicote, J. M. Montero, S. Lufti, J. Macias-Guarasa (UPM)

Austrian/Vienna German voices
Michael Pucher, Dietmar Schabus, Friedrich Neubarth (ftw.)

Finnish voices
Tuomo Raitio Johannes, Antti S Suni, Martti Vainio, Paavo Alku (TKK)
Jani K. Nurminen, Jilei Tian (Nokia)
Mikko Kurimo, Reima Karhila, Teemu Hirsimaki (TKK)

Mandarin voices
Zhenhua Ling, Yu Hu, Guo-Ping Hu (iFlytek), Yi-Jian Wu (NIT)

Japanese voices
Keiichiro Oura, Keiichi Tokuda (NIT)

Hindi voices
Joao Cabral (CSTR), B. Yegnanarayana (IIIT), S. Prasanna (IIT)

Romanian voice
Adriana Stan (Technical University of Cluj-Napoca), Matthew Aylett (Cereproc)

Ibibio voice
Moses Ekpenyong (University of Uyo), Oliver Watts


Reference: e.g.
R. Barra-Chicote, J. Yamagishi, J. M. Montero, S. King, S. Lufti, J. Macias-Guarasa
``GENERACIN DE UNA VOZ SINTTICA EN CASTELLANO BASADA EN
HSMM PARA LA EVALUACIN ALBAYZN 2008: CONVERSIN TEXTO A VOZ,''