Low-Rank Adaptation of Deep Prior Neural Networks For Room Impulse Response Reconstruction

TL;DR

This paper introduces a low-rank adaptation method for Deep Prior neural networks, enabling efficient transfer learning for room impulse response reconstruction with minimal retraining, especially when only source position changes.

Contribution

It proposes integrating Low-Rank Adaptation (LoRA) into Deep Prior models to facilitate quick adaptation to new acoustic configurations with limited data and computational effort.

Findings

LoRA enables efficient fine-tuning of Deep Prior networks.

Transfer learning with LoRA maintains high physical fidelity.

Method outperforms classical retraining in limited microphone scenarios.

Abstract

The Deep Prior framework has emerged as a powerful generative tool which can be used for reconstructing sound fields in an environment from few sparse pressure measurements. It employs a neural network that is trained solely on a limited set of available data and acts as an implicit prior which guides the solution of the underlying optimization problem. However, a significant limitation of the Deep Prior approach is its inability to generalize to new acoustic configurations, such as changes in the position of a sound source. As a consequence, the network must be retrained from scratch for every new setup, which is both computationally intensive and time-consuming. To address this, we investigate transfer learning in Deep Prior via Low-Rank Adaptation (LoRA), which enables efficient fine-tuning of a pre-trained neural network by introducing a low-rank decomposition of trainable parameters, thus allowing the network to adapt to new measurement sets with minimal computational overhead. We embed LoRA into a MultiResUNet-based Deep Prior model and compare its adaptation performance against full fine-tuning of all parameters as well as classical retraining, particularly in scenarios where only a limited number of microphones are used. The results indicate that fine-tuning, whether done completely or via LoRA, is especially advantageous when the source location is the sole changing parameter, preserving high physical fidelity, and highlighting the value of transfer learning for acoustics applications.