Schematic diagram of the lumped-element vocal fold model. Vertical gray plane represents the glottal midplane. See the text for definitions of variables.
Subglottal and supraglottal tract geometry in the voice production model. For illustration, the lower half of the plot mirrors the top half. P S = static subglottal pressure, p O (t) = acoustic pressure.
(Color online) Illustration of how lateral displacement waveforms x L and xR (of the left and right vocal folds, respectively) are derived from model outputs x 1α and x 2α (α = l, r) with Q = 0.8. Gray shading indicates times of positive glottal area.
(Color online) Parameterization of (A) lateral displacement waveforms and (B) glottal area waveforms to obtain vocal fold vibratory measures. x L , x R = lateral displacement of left/right vocal fold; OP = open phase; n L , n R = time of maximal left/right vocal fold displacement; x c , x o = mediolateral position of vocal folds at onset of glottal closure/opening; A L , A R = peak-to-peak lateral displacement of left/right vocal fold; w = maximum glottal width; PP = plateau phase; P = period; and a, b = opening/closing phase.
Effects of changing model parameter Q, in increments of 0.01, on left–right phase asymmetry PA (black) and left–right amplitude asymmetry AA (gray).
(Color online) Waveforms from model simulations with Q = 1 (left column) and Q = 0.77 (right column).
Model-based covariations of left–right phase asymmetry PA with (A) plateau quotient PQ, (B) axis shift during closure AS, (C) closing quotient ClQ, and (D) spectral tilt TL*.
(Color online) Case study of subject N1 with a low degree of left–right phase asymmetry PA (6%). Plots display the lateral displacement waveforms, phonovibrogram, glottal area waveform, radiated acoustic pressure waveform, and magnitude frequency spectra of the glottal area and radiated acoustic pressure.
(Color online) Case study of subject P13 with a high degree of left–right phase asymmetry PA (–51%). Cf. Fig. 8 .
(Color online) Scatter plots of statistically significant relationships in the subject data. (A) PA and AA, (B) PA and AS, (C) |PA| and ClQ, and (D) ClQ and TL*. Pearson’s r and p-value are indicated. Pairwise relationships obtained from the model simulations sweeping the Q parameter are overlaid on their respective plots. Note that the model trendline in (D) is derived from model data as ClQ is not an independent model parameter. Axes are linear, with tick marks positioned at the abscissa and ordinate of each point and longer tick marks (gray) at 10% or 2 dB/octave intervals.
Summary statistics of HSV-based asymmetry measures, HSV-based glottal area measures, and spectral measures of the radiated acoustic pressure waveform for human subject recordings a .
Pearson’s correlation coefficient r for significant pairwise relationships among average values of HSV-based asymmetry measures, HSV-based glottal area measures, and spectral measures of the radiated acoustic pressure waveform from all subject recordings (N = 53) a .
Article metrics loading...
Full text loading...