Hiroki Matsuzaki

Finite‐element method analysis of a three‐dimensional vocal tract model with branches

Sound wave propagation in a vocal tract model is investigated using a three‐dimensional finite‐element method (FEM). In the present study, a new FEM model is proposed with branches of the vocal tract measured by MRI. The branches correspond to the pyriform fossa and the vallecula of epiglottis. Using this vocal tract model, sound pressure and particle velocity are computed. Evaluations are as follows: first, the difference of vocal tract transfer function (VTTF) between the model with the branches and without; and, second, the difference of VTTF between the model taking into account the wall impedance of the tongue and not taking it into account. From the simulation results, it is shown that formant peak levels are influenced by the branches in the high‐frequency region over 4 kHz, and the distribution of the sound‐pressure pattern around the branches is seen by a complicated spherical surface. Furthermore, in the case of the model with the wall impedance, it is shown that the formant peak levels have a t...

3D finite element analysis of Japanese vowels in elliptic sound tube model

Based on vocal tract cross-section data formed by magnetic resonance images of the head during pronunciation of the Japanese vowels /a/, /i/, and /u/, a three-dimensional sound tube model is constructed with its cross section approximated by an equivalent ellipse with a peripheral length of the vocal tract cross section and with the same cross-sectional area. By means of the finite element method, the sound pressure and the particle velocity in the sound tube are computed. For the sound tube model, a vocal tract wall impedance modeling a soft wall is given as the boundary condition. A radiation space to represent the three-dimensional radiation at the radiation lips is attached at the radiation end. From the simulation results, the radiation impedance and the vocal tract transfer function are computed and the results are compared with the conventional one-dimensional model. The first and second formants agree well with those given by the one-dimensional model, whereas there is a difference in the bandwidth and the formant frequency at the third and higher formant frequencies due to higher-order modes in the sound tube. It is also shown that the radiation impedance is more dependent on the shape of the radiation aperture than on the shape of the sound tube and the wall boundary conditions.

FEM simulation of tongue deformation for /i/ with a four‐cube model applied to tagged cine‐MRI data

Roles of extrinsic and intrinsic tongue muscles in the production of vowel /i/ were examined using a finite element method (FEM) applied to the tagged cine‐MRI data. It has been thought that tongue tissue deformation for /i/ is mainly due to the combined actions of the genioglossus muscle bundles advancing the tongue root to elevate the dorsum with a mid‐line grooving. A recent study with the tagging‐MRI revealed earlier, faster and greater tissue deformation at anterior top of the tongue than posterior part during /ei/ production. This result implies the contribution of the intrinsic tongue muscle (transverse anterior) with an independent hydrostat mechanism from that of the genioglossus muscle bundles. In this study, a simple four‐cube model is built to examine the co‐contraction effect of the genioglossus and transverse muscles using using the FEM. The simulation result with the anterior transverse muscle (Ta) showed good agreement with the pattern of the tongue deformation obtained from the tagged‐MRI...

Finite‐element method analysis of acoustic characteristics of the vocal tract with the nasal cavity during phonation of Japanese /a/

For this study, the transfer functions and active sound intensities of a vocal tract model with and without a nasal cavity were computed using a three‐dimensional finite‐element method (FEM). The models were based on vowel data obtained by magnetic resonance imaging (MRI) of the vocal tract with a nasal cavity during phonation of the Japanese /a/. The oral cavity was also coupled with the nasal cavity in a three‐dimensional volume of radiation. Effects of wall impedance were also examined. Coupling of the nasal cavity to the oral cavity indicated the following aspects. Additional peaks appeared below 3 kHz for the lossless condition. However, they disappeared in the simulation for the soft wall condition. The sound energy circulation did not occur in the simulation for the soft wall condition. Regarding effects of the wall boundary condition on the spectral envelope, three‐dimensional simulations confirmed the upward shift of lower formant frequencies. However, disagreement of the formant frequencies betw...

Computation of the acoustic characteristics of vocal-tract models with geometrical perturbation by using higher-order modes

This work presents computational results of sound pressure distributions and transfer characteristics for a large number of vocal-tract models which contain geometrical perturbations. A small change of the vocal-tract shape is regarded as a geometrical perturbation of the axis position of each vocal-tract section. Computation of the acoustic field in the models are performed using higher-order modes. The results indicate that acoustic characteristics in the higher frequencies are highly sensitive to the small change of the vocal-tract shape.

Computation of the acoustic characteristics of simplified vocal-tract models by 3-D finite element method

The acoustic characteristics of vocal-tract models are investigated using a three-dimensional finite element method. Several finite element models are created in order to examine the influence of the simplification of the vocal-tract shapes on the finer acoustic characteristics up to the higher frequencies of around 8 kHz. The simplification takes into account the following aspects: cross-sectional shape, bend, branches of the pyriform fossa and the valleculla, and lip protrusion. The effects of the yielding vocal-tract wall are also studied. The results indicate that with the frequency range up to around the third formant, the simplification of the cross section and the bend is considered to be an effective approximation. It is shown that the use of soft walls suppresses the appearance of sharp peaks and zeros in the frequencies above 3 kHz. It is suggested that the lips and the branches independently affect the shifts of the first three formant frequencies.