English

SHAMaNS: Sound Localization with Hybrid Alpha-Stable Spatial Measure and Neural Steerer

Sound 2025-06-25 v1 Artificial Intelligence Machine Learning Audio and Speech Processing

Abstract

This paper describes a sound source localization (SSL) technique that combines an α\alpha-stable model for the observed signal with a neural network-based approach for modeling steering vectors. Specifically, a physics-informed neural network, referred to as Neural Steerer, is used to interpolate measured steering vectors (SVs) on a fixed microphone array. This allows for a more robust estimation of the so-called α\alpha-stable spatial measure, which represents the most plausible direction of arrival (DOA) of a target signal. As an α\alpha-stable model for the non-Gaussian case (α\alpha \in (0, 2)) theoretically defines a unique spatial measure, we choose to leverage it to account for residual reconstruction error of the Neural Steerer in the downstream tasks. The objective scores indicate that our proposed technique outperforms state-of-the-art methods in the case of multiple sound sources.

Keywords

Cite

@article{arxiv.2506.18954,
  title  = {SHAMaNS: Sound Localization with Hybrid Alpha-Stable Spatial Measure and Neural Steerer},
  author = {Diego Di Carlo and Mathieu Fontaine and Aditya Arie Nugraha and Yoshiaki Bando and Kazuyoshi Yoshii},
  journal= {arXiv preprint arXiv:2506.18954},
  year   = {2025}
}

Comments

European Signal Processing Conference (EUSIPCO), Sep 2025, Palermo, Italy