Learning Vocal-Tract Area and Radiation with a Physics-Informed Webster Model

TL;DR

This paper introduces a physics-informed neural network for modeling vocal-tract area and radiation in singing-voice synthesis, achieving stable and accurate spectral envelope reproduction through a Webster model-based approach.

Contribution

It develops a physics-informed Webster model trained with neural networks to estimate vocal-tract and radiation parameters, integrating PDE constraints for stable, interpretable synthesis.

Findings

Spectral envelopes are reproduced competitively with a DDSP baseline.

Model remains stable under discretization changes and pitch shifts.

Waveform quality shows breathiness, indicating areas for future improvement.

Abstract

We present a physics-informed voiced backend renderer for singing-voice synthesis. Given synthetic single-channel audio and a fund-amental--frequency trajectory, we train a time-domain Webster model as a physics-informed neural network to estimate an interpretable vocal-tract area function and an open-end radiation coefficient. Training enforces partial differential equation and boundary consistency; a lightweight DDSP path is used only to stabilize learning, while inference is purely physics-based. On sustained vowels (/a/, /i/, /u/), parameters rendered by an independent finite-difference time-domain Webster solver reproduce spectral envelopes competitively with a compact DDSP baseline and remain stable under changes in discretization, moderate source variations, and about ten percent pitch shifts. The in-graph waveform remains breathier than the reference, motivating periodicity-aware objectives and explicit glottal priors in future work.