Michael Triep 1 , Clemens Kirmse 1 , Michael Stingl 2 , Michael Döllinger 3 , Christoph Brücker 1

1 TU Bergakademie Freiberg, Institute of Mechanics and Fluid Dynamics, Freiberg

2 Universität Erlangen-Nürnberg, Institute of Applied Mathematics II, Erlangen

3 Universitätsklinikum Erlangen, Dept. of Phoniatrics and Pedaudiology, Erlangen

Clinical observations of the self-excited vocal folds in human probands during phonation allow a classification into several glottal opening modulation categories. In general, one distinguishes the healthy regular and other pathological irregular glottis closure types. A slight pathological change of the vocal fold's geometry and/or their movement may have already a strong effect on the flow field and thereby on the flow-induced noise (sources) which is a parameter that characterizes the voice quality (noise-to-harmonic ratio). In order to study the flow-dependent causes for the deterioration in voice quality, a realistic scaled-up three-dimensional (3-D) dynamic glottal cam model which has been already used in a previous study for regular glottal opening modulation situations is used. In this reference case the elliptic-shaped orifice opening cross section and a convergent - straight - divergent lateral and upward movement mimics the realistic time-varying cyclic motion of the healthy vocal folds and results in a realistic volume flow waveform profile.

Four different glottis closure types from clinical observations are mapped individually onto the membrane covered 3-D cam model considering a comparable maximum opening area for all, thus allowing us to differentiate deviations in the flow field from the reference case. The flow through the driven cam model is studied in water while keeping the most important flow parameters similar to the real situation. The advantages of this model are the enlarged time scales and the reproducible boundary conditions for cyclic measurements and visualizations of the flow. The large and small scale 3-D flow structures can be studied and compared in detail with numerical simulations. These aim to determine the harmonic and the noisy components of the flow-induced local source terms and the global noise-to-harmonic ratio, which will give insight to the mechanisms leading to deteriorated voice quality.

Willy Mattheus 1 , Stefan Zörner 2 , Manfred Kaltenbacher 2 , Rüdiger Schwarze 1 , Christoph Brücker 1 , Willy Mattheus 1

1 TU Bergakademie Freiberg, Institute of Mechanics and Fluid Dynamics, Freiberg

2 Alpen-Adria Universität Klagenfurt, Institute of Smart System-Technologies, Klagenfurt

For a three dimensional (3-D) model of the human vocal folds the time dependent transglottal flow field is computed numerically by solving the discretized incompressible Navier-Stokes equation. The model geometry is based on an existing up-scaled experimental 3-D dynamic glottal configuration for flow visualization purposes under reproducible conditions. The fluid used in the model is water. The flow is pressure driven while keeping the most important flow parameters similar to the real situation in human larynx. Typical glottal opening modulations observed in clinical studies and which have been mapped onto the membrane covered cams in the corresponding experimental set-up, are prescribed to the computational grid. The flow boundary conditions for the numerical simulation in the form of volume flow rate at the inlet and constant pressure over the outlet are set as in experiment. The flow field is computed with the finite volume CFD code OpenFOAM on parallel computer architecture. The flow downstream of several irregular glottal closure cases is studied and compared to the flow in the regular case. In addition, the 3-D time-dependent vortical flow structures are compared to experimental flow results. The coherent large-scale structures and stochastic small-scale structures are examined in more detail in order to determine their contribution to the global sound pressure spectrum. Based on Lighthill's acoustic analogy, the acoustic source term distributions and the sound propagation are numerically computed with the finite element code CFS . Since, in general, the quality of voice depends on the ratio of the noisy to the harmonic components; results from the present study are intended to help surgeons in the diagnostics and therapy of pathological vocal folds.

Ken-Ichi Sakakibara 1 , Hiroshi Imagawa 2 , Hisayuki Yokonishi 2 , Takao Goto 3 , Miwako Kimura 3,4 , Takaharu Nito 2 , Niro Tayama 3

1 Health Sciences Univ., Dept. Comm. Disorders, Sapporo

2 University of Tokyo, Dept. Otolaryngology, Tokyo

3 Intenational Medical Center of Japan, , Tokyo

4 UT Southwestern Medical Center, Dept. Otolaryngology, Dallas

Inspiratory phonation is the phonation during inhalation and in
general, it sometimes appear in laughing.  In this study, we analyze
vibratory patterns of reverse phonation using stereographic high-speed
imaging, EGG, and simulation by two-mass model with the supra- and
subglottal system.  In reverse phonation, mucosal waves of the vocal
folds were observed in the high-speed images and propagated from the
top to the bottom (from lateral to center). The vocal fold oscillation
is easy in fry and falsetto registers. In modal register, the vocal
fold osicllation is often unstable in human and simulation.

Jaromír Horáček 1 , Václav Uruba 1 , Vojtěch Radolf 1 , Vaclav Uruba 1 , Petr Šidlof 1 , Jan Veselý 1 , Vítězslav  Bula 1

1 Institute of Thermomechanics, Academy of Sciences, Prague

The contribution describes a complex physical model of the voice production that consists of simplified 1:1 scaled models of the trachea, the self-oscillating vocal folds and the vocal tract with acoustical spaces that correspond to the vowel /a:/. A simplified “2D” plexiglass model of the vocal tract was developed from the 3D FE models designed from MR images. The vocal folds were substituted by a model made of a latex thin cover layer filled by polyurethane rubber. The vocal fold model was joined to the model of the subglottal spaces. The measurement set-up enabled to use the PIV method for visualization of the airflow and to perform measurements of dynamic subglottal pressure, radiated acoustic pressure and vocal fold vibration. The time-resolved PIV method was used for instantaneous velocity field evaluation. The measuring system DANTEC consisted of laser with maximum frequency 10 kHz (depending on the field of view) and the camera by which 1000 consecutive snaps were acquired and evaluated. Preliminary results will be presented for measurements performed within a physiologically real range of input parameters (mean air flow rate Qmean=0.26-0.6 l/s, fundamental frequency F0=150-172 Hz). Ten images of the vibrating vocal folds during one oscillation period were recorded by the high-speed camera in the same instants as the PIV velocity fields. In a wider region above the ventricular folds it is possible to see large vortices with dimensions comparable with the channel cross-section. The vortices disappear in the narrowest epilaryngeal part of the vocal tract where the flow is uniform. Large eddies of a size comparable with the channel height were observed in the model of the mouth cavity. Due to Coanda effect, the airflow was attached to one wall of the channel. The airflow velocity and streamline patterns measured for two values of the mean airflow rate Qmean=0.25 l/s (Psub=900 Pa, F0=158 Hz) and Qmean=0.5 l/s (mean subglottal pressure Psub=1600 Pa, F0=168 Hz) are compared. For space-time analysis of coherent structures in the flow the bi-orthogonal decomposition method is used. The flow field is decomposed into several types of “modes”. The first most energetic mode characterizes a mean flow pattern, a few other modes a quasi-periodic structure of the flow field. Higher modes characterize mainly random disturbances in the flow. The research is supported by the Grant Agency of the Czech Republic by project No 101/08/115 “Computer and physical modelling of vibroacoustic properties of human vocal tract for optimization of voice quality”.