Articulatory synthesis from X-rays and inversion for an adaptive speech robot

摘要

Describes a speech robotic approach to articulatory synthesis. An anthropomorphic speech robot has been built, based on a real reference subject's data. This robot, called the Articulotron, has a set of relevant degrees of freedom for speech articulators, jaw, tongue, lips and larynx. The associated articulatory model has been elaborated from cineradiographic midsagittal profiles recorded in synchrony with front lip views; the model of the noise source for fricative excitation has been derived from acoustic and aerodynamic measurements on the same reference subject. In a first phase, the Articulotron has been used to perform the copy synthesis of the vowels, fricative and plosive consonants in the X-ray corpus. This allows one to assess the performance of the Articulotron in producing fairly high-quality speech, and provides a reference against which other attempts at articulatory synthesis can be compared. In a second phase, the Articulotron has be used to recover articulatory gestures from audio-visual speech prototypes. A gradient descent algorithm is used to learn the articulatory trajectories of the robot by optimisation, starting from the formant trajectories and the knowledge of constraints for the consonantal constriction or closure, in order to mimic the original VCV (vowel-consonant-vowel) audio-visual sequences. The adaptive skill of the robot is demonstrated through articulator perturbation experiments and through the elaboration of relevant strategies in the hyper/hypo-speech paradigm.