Bulletin of the American Physical Society
76th Annual Meeting of the Division of Fluid Dynamics
Sunday–Tuesday, November 19–21, 2023; Washington, DC
Session R14: Biofluids: Phonation and Speech |
Hide Abstracts |
Chair: Xudong Zheng, Rochester Institute of Technology Room: 144AB |
Monday, November 20, 2023 1:50PM - 2:03PM |
R14.00001: Computational modeling and experiment validation of vocal fold vibration for type-1 thyroplasty Amit Avhad, Zheng Li, Lea Sayce, Azure Wilson, Bernard Rousseau, Haoxiang Luo We present an integrated computational and experiment study of type-1 thyroplasty (TP) and its impact on vocal fold (VF) vibration. TP is a common intervention for unilateral VF paralysis (UVFP), involving medialization of the paralyzed VF via an implant. In the ex vivo experiment, a rabbit larynx was used to for the study, where “healthy” side of VF was medialized via trans-muscular suture and the “paralyzed size” was medialized by implant insertion. Flow experiment was then done to induce the VF vibration. The computational model was based on the pre-operative MRI scan, and the FEM model of the VF was numerically implanted to achieve the medialized configuration. This configuration was then used to perform fluid-structure interaction (FSI) simulation of the VF vibration to assess the effect of the implant. The FEM simulation of the medialization was validated against the post-operative MRI scan by comparing the VF displacement; and the FSI simulation was validated against the phonation experiment by comparing the vibratory characteristics of the VF against the high-speed video. Furthermore, the flow characteristics showed evidence of restored pulsatile glottal airflow. These results support the use of current computational model for pre-surgical planning of type-1 thyroplasty. |
Monday, November 20, 2023 2:03PM - 2:16PM |
R14.00002: Abstract Withdrawn
|
Monday, November 20, 2023 2:16PM - 2:29PM |
R14.00003: The air we breathe when speaking: links between language and fluid mechanics Junshi Wang, Simon Mendez, Haibo Dong, Manouk Abkarian, Howard A Stone Clear evidence shows that airflow jets produced when we speak contribute to the transport of airborne pathogens, such as the SARS-CoV-2 virus. In linguistics, there are clear distinctions in the manners of articulation and airstream mechanisms between vowels and consonants. However, it is still elusive how the linguistic features of vowels and consonants may determine the fluid mechanics features of the airflow jets that accompany speech. We combine experimental and numerical approaches to investigate the flow patterns produced by representative vowels—/a, i, o, u, ə/, and consonants—/p, k, f, s, h/. A 3D vocal tract is modeled with a temporally varying exit that captures key morphologic and kinematic features of the human vocal tract, including teeth and lips. An incompressible flow solver based on a sharp-interface immersed-boundary-method (IBM) is employed to compute the resultant airflow. By examining various combinations of vowels and consonants, we show that the combined effort of a large volumetric flow rate dominated by consonants and high flow speed determined by vowels contribute to faster and longer penetrations of airflow jets. We also show relative position between lips and teeth determines the orientation of the jets. This work helps bring insights into the understanding of articulatory phonetics, and the links to different languages, from a fluid mechanics perspective. |
Monday, November 20, 2023 2:29PM - 2:42PM |
R14.00004: Phase-averaged analysis of jet dynamics in a scaled up vocal fold model with asymmetric motions Timothy Wei, Abigail Haworth, Nathan Wei, Hunter Ringenberg, Michael H Krane This study focuses on the effects of glottal jet dynamics on phonation when one of the vocal folds does not move as much as the other. This can be a pathological condition in which a vocal fold is completely or partially paralyzed. Experiments were conducted using a 10x scaled-up model in a free surface water tunnel. 2-D vocal fold models with semi-circular ends were computer driven inside a square duct with constant opening and closing speeds. Four cases were studied in which one vocal fold moved 0%, 50%, 75%, and 100% of the other; the last case being, of course, the nominally 'healthy', symmetric case. Time resolved DPIV and pressure measurements along the duct centerline were made at Re = 7200 at a reduced frequency of 0.0261, corresponding to an equivalent life frequency of 97.5 Hz. Phase-averaged analysis of key contributors to sound production was conducted using terms in the streamwise integral momentum equation. The goal was to try to decouple effects associated with varying maximum gap opening, asymmetry due to partial motion of one vocal fold, and pseudo-frequency effects arising when the two vocal folds move at different speeds. Implications on energetics and sound quality are explored. |
Monday, November 20, 2023 2:42PM - 2:55PM |
R14.00005: Numerical simulation of aeroacoustic sound generation during tongue articulation and velopharyngeal closing Tsukasa Yoshinaga, HsuehJui Lu, ChungGang Li, Kazunori Nozaki, Akiyoshi Iida, Makoto Tsubokura The velopharyngeal air leakage to a nasal cavity is known to cause problematic sound generation for oral consonant pronunciation. In this study, effects of the velopharyngeal closing ratio during tongue articulation of fricative [s] on the sound generation are investigated by compressible flow simulations. The vocal tract geometries of /usu/ were extracted from a four-dimensional computed tomography scan of a subject who has no speech disorder, and the articulation process was simulated from the end of /u/ to the middle of /s/. The opening ratio of velopharyngeal valve was changed for each simulation. The turbulent airflow near the tongue constriction was modeled with the large eddy simulation, and sound propagation was directly computed by solving compressible Navier-Stokes equations. The moving immersed boundary method for the hierarchical structured grid was adopted to reduce the computational costs. When the velum and tongue moved normally as measured with the subject, the co-articulation process between /u/ and /s/ was observed, and the acoustic prediction agreed with the measurement. When the velum was not closed with the tongue elevation, the sound amplitude significantly decreased, suggesting the capability of the simulation to be utilized for the prediction of velopharyngeal insufficiency. |
Monday, November 20, 2023 2:55PM - 3:08PM |
R14.00006: Assessment of Physiological Parameters Influencing Oxygenation of Vocal Fold Tissue Rana Zakerzadeh, Isabella McCollum Phonation results from fluid-structure interactions (FSI) between the glottal airflow and the poroelastic tissue of the vocal folds (VFs). In our previous research, a poroelastic FSI model was developed to study the interstitial fluid flow dynamics in vibrating VFs and local changes in perfusion within the tissue, which are commonly associated with voice disorders such as vocal fatigue and dehydration. In this research, we aim to combine porous VFs with a mass transport model to investigate the association of blood flow with oxygen supply that corresponds with dysfunctions like hypoxia. A multiphysics computational framework by considering unsteady Navier-Stokes equations for airflow, Brinkman equation for porous VFs, and advection-diffusion-reaction equation for oxygen flow is developed. Previous experimental observations report contradictory relationships regarding VF oxygen transport and suggest that other physiological conditions, such as reaction rate, subglottal lung pressure, and permeability, may influence VF oxygenation. To evaluate the potential influences, simulations by variation of these parameters are performed and the filtration velocity and oxygen concentration are measured and compared. The outcomes highlight the importance of poroelasticity in phonation models. |
Monday, November 20, 2023 3:08PM - 3:21PM |
R14.00007: A hybrid physics informed neural network model for patient specific phonation simulation Xudong Zheng, Biao Geng, Xin-yang Liu, Jian-Xun Wang, Qian Xue This research aims to develop a new AI-enabled data-assimilation computational framework that enables seamless integration of multimodal experimental/clinical data and high-fidelity subject-specific modeling of human/animal vocal systems to provide accurate, realistic, robust, efficient and reliable simulations of individual vocal system. Toward this, we designed a novel hybrid physics informed neural network(PINN) based differentiable learning algorithm that integrates a recurrent neural network model of 3D continuum soft tissue with a differentiable fluid solver to infer the 3D flow-induced vocal dynamics and other physical quantities from high speed videoendoscopy. The effectiveness and merit of the proposed algorithm is demonstrated in subject-specific voice production problems by using synthetic data from a canine VF model and in-vivo experimental data of pigeon VFs. Results revealed that the algorithm successfully reconstructed the full three-dimensional motion of vocal fold, as well as estimation of other features such as flow rate and acoustic signals, which are difficult to be measured experimentally. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700