Dataset enables estimating room shapes and materials from sounds
ARPEGE: Dataset and Methods to Estimate Acoustic and Geometric Parameters of Rooms from Sound
Sound
Summary
People wonder if you can tell the shape and materials of a room just by listening to sounds in it. The authors created ARPEGE, a collection of recordings made in empty rooms with different sizes and wall materials, capturing how sounds bounce around. They also included information about where sound sources and microphones were placed, along with some human speech and handclaps. They tested two advanced methods that guess room features from these recordings. They provide tools to recreate the sounds digitally, aiming to make a detailed digital copy of the rooms.
What this means in practice
- •For audio engineers: Use the dataset and methods to calibrate room acoustics for better sound recording and reproduction setups.
- •For virtual reality developers: Create more realistic indoor audio environments by digitally modeling rooms from recorded sound reflections.$Commercial implications: Enables selling immersive VR applications with accurate room acoustics and spatial audio simulation.
Authors
Jean-Daniel Pascal Prieto, Prerak Srivastava, Cédric Foy, Antoine Deleforge
Abstract
Can one hear the walls of a room? To foster audio signal processing research around this question we release ARPEGE, a dataset containing hundreds of ambisonic spherical-array room impulse responses recorded in 6 empty rooms, annotated by the relative positions of sources, microphones and all acoustic reflectors. The dataset includes multiple room sizes, surface materials, orientations and distances, as well as recordings of human speakers and handclaps. Two state-of-the-art methods on image source localization and blind room parameter estimation are demonstrated on the data. Initial code for re-simulating the measured room impulse responses is provided, a longer term goal being to digitally twin the dataset.