Evaluation of Neural Surrogates for Physical Modelling Synthesis of Nonlinear Elastic Plates
This work addresses the computational bottleneck in real-time audio synthesis for applications like music production, but it is incremental as it focuses on evaluating existing methods rather than introducing new ones.
The paper tackled the problem of real-time audio synthesis from physical simulations of nonlinear elastic plates by evaluating neural network surrogates, finding that current models have limitations in long-sequence prediction despite being trained on short sequences.
Physical modelling synthesis aims to generate audio from physical simulations of vibrating structures. Thin elastic plates are a common model for drum membranes. Traditional numerical methods like finite differences and finite elements offer high accuracy but are computationally demanding, limiting their use in real-time audio applications. This paper presents a comparative analysis of neural network-based approaches for solving the vibration of nonlinear elastic plates. We evaluate several state-of-the-art models, trained on short sequences, for prediction of long sequences in an autoregressive fashion. We show some of the limitations of these models, and why is not enough to look at the prediction error in the time domain. We discuss the implications for real-time audio synthesis and propose future directions for improving neural approaches to model nonlinear vibration.