Nonlinear dynamics of phonations in excised larynx experiments

Area of Science:

• * Voice science and bioacoustics
• * Nonlinear dynamics and chaos theory

Background:

• * Irregular voice phonations exhibit aperiodic waveforms and broadband spectra.
• * Distinguishing irregular from normal phonations is crucial for diagnosing voice disorders.

Purpose of the Study:

• * To quantitatively analyze phonations in excised larynx experiments using nonlinear dynamic methods.
• * To assess the efficacy of correlation dimension, Lyapunov exponents, jitter, shimmer, and peak prominence ratio in differentiating irregular from normal phonations.

Main Methods:

• * Application of nonlinear dynamic methods: correlation dimension and Lyapunov exponents.
• * Statistical analysis to compare nonlinear measures and traditional voice perturbation measures (jitter, shimmer, peak prominence ratio).

Main Results:

• * Finite correlation dimensions and positive Lyapunov exponents confirm chaos in irregular phonations.
• * Correlation dimension and maximal Lyapunov exponent significantly differentiate irregular from normal phonations.
• * Jitter, shimmer, and peak prominence ratio did not show significant differences between phonation types.

Conclusions:

• * Nonlinear dynamic methods, specifically correlation dimension and Lyapunov exponents, are effective tools for identifying chaos in voice production.
• * These methods offer a more robust approach than traditional measures for distinguishing irregular from normal phonations.
• * Findings support the development of advanced diagnostic tools for voice disorders.