A Study of Vocal Fold Vibration Using a Slightly Compressible Fluid Domain

During human voice production, air forced from the lungs through the larynx induces vibration of the vocal folds. Computational models of this coupled fluid-solid system have traditionally utilized an incompressible fluid domain. However, studies have shown that coupling of tracheal acoustics with vocal fold dynamics is significant. Further, in the absence of compressibility, some models fail to achieve self-sustained vibration. This presentation discusses a slightly compressible airflow model, fully coupled with a vocal fold tissue model, as a possible substitute for the traditional incompressible approach. The derivation and justification of the slightly compressible fluid model are discussed. Results are reported of a study of the nature of the coupling between the fluid and vocal fold regions for both slightly compressible and incompressible fluid domains using a commercial fluid-solid finite element package. Three different types of inlet boundary conditions, including constant pressure, constant velocity, and moving wall, are explored. The incompressible and slightly compressible models with the three boundary conditions are compared with each other and with experimental data obtained using synthetic self-oscillating vocal fold models. The results are used to validate the slightly compressible flow model as well as to explore candidate boundary conditions for vocal fold vibration simulations.