CIFASIS 20631
CENTRO INTERNACIONAL FRANCO ARGENTINO DE CIENCIAS DE LA INFORMACION Y DE SISTEMAS
Unidad Ejecutora - UE
congresos y reuniones científicas
Título:
Language Independent Speech-driven Facial Animation System
Autor/es:
LUCAS D. TERISSI, JUAN CARLOS GÓMEZ
Lugar:
Santa Fe, Argentina
Reunión:
Conferencia; XXXIV Conferencia Latinoamericana de Informática (CLEI 2008); 2008
Institución organizadora:
SADIO
Resumen:
In this paper, a speech-driven MPEG-4 compliant facial animation system is proposed. The system is based on the inversion of a joint Audio-Visual Hidden Markov Model to estimate the visual information from speech data. The inversion algorithm is derived for the general case of considering full covariance matrices for the audio-visual observations. The proposed speech-driven facial animation system operates at a subphonemic level, with the advantage of being adaptable to any language. The experiments are carried out using audio-visual databases compiled by the authors.
