Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Speakers Localization Using Batch EM In Unfolding Neural Network

Rina Veler, Sharon Gannot

Abstract

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.

Speakers Localization Using Batch EM In Unfolding Neural Network

Abstract

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.
Paper Structure (5 sections, 12 equations, 1 figure, 1 table)

This paper contains 5 sections, 12 equations, 1 figure, 1 table.

Table of Contents

  1. Introduction
  2. Problem Formulation
  3. Proposed Method
  4. Experimental Study
  5. Conclusion

Figures (1)

  • Figure 1: Batch EM unfolding neural network architecture.