Reciprocal Latent Fields for Precomputed Sound Propagation

TL;DR

This paper introduces Reciprocal Latent Fields, a memory-efficient framework for encoding and predicting acoustic parameters in sound propagation, achieving high fidelity with significantly reduced memory requirements for real-time virtual environments.

Contribution

The paper presents RLF, a novel reciprocal latent encoding method that improves memory efficiency and acoustic realism in precomputed sound propagation simulations.

Findings

RLF reduces memory usage by several orders of magnitude.

Sound rendered with RLF is perceptually indistinguishable from ground-truth.

RLF effectively models complex acoustic phenomena in large scenes.

Abstract

Realistic sound propagation is essential for immersion in a virtual scene, yet physically accurate wave-based simulations remain computationally prohibitive for real-time applications. Wave coding methods address this limitation by precomputing and compressing impulse responses of a given scene into a set of scalar acoustic parameters, which can reach unmanageable sizes in large environments with many source-receiver pairs. We introduce Reciprocal Latent Fields (RLF), a memory-efficient framework for encoding and predicting these acoustic parameters. The RLF framework employs a volumetric grid of trainable latent embeddings decoded with a symmetric function, ensuring acoustic reciprocity. We study a variety of decoders and show that leveraging Riemannian metric learning leads to a better reproduction of acoustic phenomena in complex scenes. Experimental validation demonstrates that RLF maintains replication quality while reducing the memory footprint by several orders of magnitude. Furthermore, a MUSHRA-like subjective listening test indicates that sound rendered via RLF is perceptually indistinguishable from ground-truth simulations.