Extraction of vocal tract area function from three-dimensional magnetic resonance images using digital waveguide mesh
Kenji Inoue, Hironori Takemoto, Tatsuya Kitamura, Shinobu Masaki, Hirotake Nakashima · The Journal of the Acoustical Society of America · 2008
A method is proposed in this paper to extract the vocal tract area function from the three-dimensional magnetic resonance images. The proposed method uses the digital waveguide mesh, an implementation of the finite-difference time-domain (FDTD) method, to simulate wave propagation in the vocal tract from the glottis to the lips. The dimensions of the vocal tract areas are then calculated along the traveling wavefront that emerges from the simulation. Formant analysis has been conducted for Japanese vowels to show the validity of the proposed method. The calculated formant frequencies of the area functions obtained by the proposed method and other existing methods have been compared to those measured from the acoustic utterance of the imaged person.