Generative adversarial networks with physical sound field priors

TL;DR

This paper introduces a GAN-based method utilizing physical sound priors for accurate, flexible, and high-frequency sound field reconstruction from limited measurements, outperforming existing techniques.

Contribution

It presents a novel deep learning approach that incorporates physical priors into GANs for improved sound field reconstruction from sparse data.

Findings

Enhanced accuracy in high-frequency reconstruction

Effective extrapolation beyond measurement regions

Robust performance across different measurement setups

Abstract

This paper presents a deep learning-based approach for the spatio-temporal reconstruction of sound fields using Generative Adversarial Networks (GANs). The method utilises a plane wave basis and learns the underlying statistical distributions of pressure in rooms to accurately reconstruct sound fields from a limited number of measurements. The performance of the method is evaluated using two established datasets and compared to state-of-the-art methods. The results show that the model is able to achieve an improved reconstruction performance in terms of accuracy and energy retention, particularly in the high-frequency range and when extrapolating beyond the measurement region. Furthermore, the proposed method can handle a varying number of measurement positions and configurations without sacrificing performance. The results suggest that this approach provides a promising approach to sound field reconstruction using generative models that allow for a physically informed prior to acoustics problems.