Differentiable Modal Synthesis for Physical Modeling of Planar String Sound and Motion Simulation

TL;DR

This paper presents a neural network-based model for simulating nonlinear string vibrations that integrates physical laws with spectral and modal synthesis, improving accuracy in sound and motion simulation.

Contribution

It introduces a novel differentiable model combining modal synthesis and spectral modeling for physically accurate nonlinear string simulation.

Findings

Achieves superior accuracy over baseline models in string motion simulation.

Successfully integrates physical properties into neural network architecture.

Provides open-source code and demo for reproducibility.

Abstract

While significant advancements have been made in music generation and differentiable sound synthesis within machine learning and computer audition, the simulation of instrument vibration guided by physical laws has been underexplored. To address this gap, we introduce a novel model for simulating the spatio-temporal motion of nonlinear strings, integrating modal synthesis and spectral modeling within a neural network framework. Our model leverages physical properties and fundamental frequencies as inputs, outputting string states across time and space that solve the partial differential equation characterizing the nonlinear string. Empirical evaluations demonstrate that the proposed architecture achieves superior accuracy in string motion simulation compared to existing baseline architectures. The code and demo are available online.