A Physics-Informed Neural Network-Based Approach for the Spatial Upsampling of Spherical Microphone Arrays

2024/07/26 by Miotello, Federico, Terminiello, Ferdinando, Pezzoli, Mirco +3 · 2 citations
#Audio and Speech Processing (eess.AS) #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Signal Processing (eess.SP) #Sound (cs.SD) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2407.18732

Abstract

Spherical microphone arrays are convenient tools for capturing the spatial characteristics of a sound field. However, achieving superior spatial resolution requires arrays with numerous capsules, consequently leading to expensive devices. To address this issue, we present a method for spatially upsampling spherical microphone arrays with a limited number of capsules. Our approach exploits a physics-informed neural network with Rowdy activation functions, leveraging physical constraints to provide high-order microphone array signals, starting from low-order devices. Results show that, within its domain of application, our approach outperforms a state of the art method based on signal processing for spherical microphone arrays upsampling.

Cited by

• Kernel ridge regression based sound field estimation using a rigid spherical microphone array
• Dynamic Real-Time Ambisonics Order Adaptation for Immersive Networked Music Performances

Related

• Circumvent spherical Bessel function nulls for open sphere microphone arrays with physics informed neural network
• Neural Ambisonics encoding for compact irregular microphone arrays
• Spatial Upsampling of Head-Related Transfer Functions Using a Physics-Informed Neural Network
• A circular microphone array with virtual microphones based on acoustics-informed neural networks
• Gen-A: Generalizing Ambisonics Neural Encoding to Unseen Microphone Arrays