Physics-informed neural network for predicting in vacuo vocal fold eigenmodes: A proof of concept study

Al Khasawneh MEM, Döllinger M, Zhang Z (2026)

Publication Type: Journal article

Publication year: 2026

Journal

JASA Express Letters Acoustical Society of America (ASA)

Book Volume: 6

Article Number: 045202

Journal Issue: 4

DOI: 10.1121/10.0043248

Abstract

This study investigates a machine-learning approach for real-time computation of in vacuo vocal fold eigenmodes. A physics-informed neural network is trained to predict eigenmodes and eigenfrequencies by integrating the governing equations of vocal fold dynamics. The proposed architecture predicts physically consistent modal shapes and frequency estimates, showing strong agreement with finite-element method results for lower-order modes, achieving mean relative errors below 6% in eigenfrequency prediction and cosine correlation values approaching 1 in eigenmode prediction, while prediction accuracy declines for higher-order modes. These findings demonstrate that the physics-informed neural network provides an accurate and efficient real-time computation of in vacuo vocal fold eigenmodes.

Authors with CRIS profile

Michael Döllinger Professur für Computational Medicine

Involved external institutions

University of California Los Angeles (UCLA)

United States (USA) (US)

How to cite

APA:

Al Khasawneh, M.E.M., Döllinger, M., & Zhang, Z. (2026). Physics-informed neural network for predicting in vacuo vocal fold eigenmodes: A proof of concept study. JASA Express Letters, 6(4). https://doi.org/10.1121/10.0043248

MLA:

Al Khasawneh, Mohd Ethar M., Michael Döllinger, and Zhaoyan Zhang. "Physics-informed neural network for predicting in vacuo vocal fold eigenmodes: A proof of concept study." JASA Express Letters 6.4 (2026).

BibTeX: Download