Data-driven 3D Room Geometry Inference with a Linear Loudspeaker Array and a Single Microphone

TL;DR

This paper introduces a data-driven method for inferring 3D room geometry using a linear loudspeaker array and a single microphone, leveraging neural networks and acoustic beamforming to accurately localize reflectors.

Contribution

It proposes a novel supervised deep learning approach that generalizes well to real RIRs, eliminating the need for semi-supervised intermediate steps in room geometry inference.

Findings

Achieves accuracy comparable to baseline model-driven methods

Generalizes effectively to unseen RIRs

Provides a fully automated RGI framework

Abstract

Knowing the room geometry may be very beneficial for many audio applications, including sound reproduction, acoustic scene analysis, and sound source localization. Room geometry inference (RGI) deals with the problem of reflector localization (RL) based on a set of room impulse responses (RIRs). Motivated by the increasing popularity of commercially available soundbars, this article presents a data-driven 3D RGI method using RIRs measured from a linear loudspeaker array to a single microphone. A convolutional recurrent neural network (CRNN) is trained using simulated RIRs in a supervised fashion for RL. The Radon transform, which is equivalent to delay-and-sum beamforming, is applied to multi-channel RIRs, and the resulting time-domain acoustic beamforming map is fed into the CRNN. The room geometry is inferred from the microphone position and the reflector locations estimated by the network. The results obtained using measured RIRs show that the proposed data-driven approach generalizes well to unseen RIRs and achieves an accuracy level comparable to a baseline model-driven RGI method that involves intermediate semi-supervised steps, thereby offering a unified and fully automated RGI framework.