Numerical Simulation of Flow-induced Sound in Human Voice Production

The aeroacoustic mechanisms in human voice production are complex coupled processes, which are not fully understood until now. The paper presents a hybrid numerical approach to the solution of the acoustic field in human vocal tract, when the fluid flow is first solved by a dedicated CFD solver on a fine computational mesh covering the larynx. Subsequently, the acoustic sources and propagation of the sound waves are calculated by Lighthill's analogy or acoustic perturbation equations on a coarse mesh covering the whole vocal tract including radiation region. The CFD simulation was run in parallel on a SGI Altix UV shared-memory supercomputer; strong and weak parallel performance was tested for up to 50 computational cores. The spectra of the radiated sound for two vowels/i/and/u/evaluated from the acoustic simulation show good agreement with the formant frequencies known from human subjects.