Unsupervised Acoustic Scene Mapping Based on Acoustic Features and Dimensionality Reduction
This addresses acoustic scene mapping for microphone arrays, offering an incremental improvement over classical TDOA-based methods by reducing sensitivity to noise and reverberation.
The paper tackles the problem of acoustic scene mapping by introducing an unsupervised data-driven approach using local conformal autoencoders (LOCA) to learn representations isometric to spatial locations, demonstrating robustness to reverberation in simulations.
Classical methods for acoustic scene mapping require the estimation of time difference of arrival (TDOA) between microphones. Unfortunately, TDOA estimation is very sensitive to reverberation and additive noise. We introduce an unsupervised data-driven approach that exploits the natural structure of the data. Our method builds upon local conformal autoencoders (LOCA) - an offline deep learning scheme for learning standardized data coordinates from measurements. Our experimental setup includes a microphone array that measures the transmitted sound source at multiple locations across the acoustic enclosure. We demonstrate that LOCA learns a representation that is isometric to the spatial locations of the microphones. The performance of our method is evaluated using a series of realistic simulations and compared with other dimensionality-reduction schemes. We further assess the influence of reverberation on the results of LOCA and show that it demonstrates considerable robustness.