We propose an interpretable Batch-EM Unfolded Network for robust speaker localization. By embedding the iterative EM procedure within an encoder-EM-decoder architecture, the method mitigates initialization sensitivity and improves convergence. Experiments show superior accuracy and robustness over the classical Batch-EM in reverberant conditions.
@article{arxiv.2603.16278,
title = {Speakers Localization Using Batch EM In Unfolding Neural Network},
author = {Rina Veler and Sharon Gannot},
journal= {arXiv preprint arXiv:2603.16278},
year = {2026}
}