Aeroacoustic sound generation during phonation in a synthetic larynx model

Abstract

The fluid flow within the human larynx plays an essential role in the fluid-structure-acoustic interaction during voice production. This study addresses the supraglottal flow field downstream of aerodynamically driven, synthetic vocal folds based on the M5 model. The larynx replica is designed to provide full optical access to the flow region. Two different approaches based on Particle-Image-Velocimetry (PIV) were applied for measuring the flow: Phase-averaged (PA) and High-speed (HS) PIV. Beside a comparison of the supraglottal flow field, the acoustic sources were calculated based on both PIV approaches. Furthermore, the simulated far-field sound based on the HS-PIV data is compared to experimental results. Within both PIV approaches, the typical asymmetric jet flow was detected. However, transient flow field characteristics as high vorticity and maximal velocities peaks could only be observed in HS-PIV data. A strong aeroacoustic source was found immediately downstream of the glottis for both PIV approaches. However, the sources in the jet region could only be observed in the HS-PIV data owing to the averaging procedure for PA-PIV. The comparison between simulated and measured sound spectra revealed good agreement. In contrast, subharmonic tones could not be detected hinting to an additional non-aeroacoustic mechanism of sound generation.