Acoustic Reflectors Localization from Stereo Recordings Using Neural Networks

Abstract

Acoustic room geometry estimation is often performed in ad hoc settings, i.e., using multiple microphones and sources distributed around the room, or assuming control over the excitation signals. We propose a fully convolutional network (FCN) that localizes reflective surfaces under the relaxed assumptions that (i) a compact array of only two microphones is available, (ii) emitter and receivers are not synchronized, and (iii) both the excitation signals and the impulse responses of the enclosures are unknown. Our FCN is trained in a supervised fashion to predict the likelihood of reflective surfaces at specific distances and directions-of-arrival (DOA). When a single reflective surface is present, up to 80% of real and virtual sources are detected, while this figure approaches 50% in rectangular rooms. Experiments on real-world recordings report similar accuracy as with artificially reverberated speech signals, validating the generalization capabilities of the framework.

A Survey of Sound Source Localization with Deep Learning Methods

“…A lot of works consider only one source to localize, as it is the simplest scenario to address, e.g. [45], [46], [47]. We refer to this scenario as single-source SSL.…”

“…In [46], the neural network architecture includes a set of 2D convolutional layers for frame-wise feature extraction, followed by several 1D convolutional layers in the time dimension for temporal aggregation. In [69], 3D convolutional layers are applied on SRP-PHAT power maps computed for both azimuth and elevation estimation.…”

“…In [106], the azimuth axis is classified with I = 72 classes along with the distance and height of the source, but those last two quantities are classified into a very small set of possible pairs: (30,30), (90,30) and (90, 90) (in centimeters). In [46], Bologni et al trained a CNN to classify a single-source signal into a 2D map representing the azimuth and distance dimensions.…”