The generation and propagation of the human voice is studied using direct numerical simulation. A full body domain is employed for the purpose of directly computing the sound in the region past the speaker's mouth. The air in the vocal tract is modeled as a compressible and viscous fluid interacting with the vocal folds (VFs). The vocal fold tissue material properties are multi-layered, with varying stiffness, and a finite-strain model is utilized and implemented in a quadratic finite element code. The fluid-solid domains are coupled through a boundary-fitted interface and utilize a Poisson equation-based mesh deformation method. The domain includes an anatomically relevant vocal tract geometry, either in two dimensions or in three dimensions. Adult and two-year-old child anatomy inspired simulations are performed. Phonation simulations using a non-linear hyper elastic, linear elastic and viscoelastic models of the VFs are performed and compared. The sensitivity of phonation to the VF Poisson's ratio is also evaluated. Simulations are employed to investigate voice disorders related to vocal fold stiffness asymmetry and unilateral vocal fold paralysis (UVFP). Additionally, an analysis is performed for medialization laryngoplasty, a well known surgical treatment for UVFP. Phonation onset is determined from all the simulations as a measure of degree of voice disorder with phonation threshold pressure (PTP) as a key parameter for the quantification. The computational model developed is demonstrated to be consistent with prior measurements and sufficiently sensitive to be used in future studies involving VF pathologies, surgical procedures to restore voice, and/or closed loop models of voice, speech and perception.
PDF
download
878987797
028-85220240
