English

Microphone Array Signal Processing and Deep Learning for Speech Enhancement

Audio and Speech Processing 2025-01-14 v1 Sound Signal Processing

Abstract

Multi-channel acoustic signal processing is a well-established and powerful tool to exploit the spatial diversity between a target signal and non-target or noise sources for signal enhancement. However, the textbook solutions for optimal data-dependent spatial filtering rest on the knowledge of second-order statistical moments of the signals, which have traditionally been difficult to acquire. In this contribution, we compare model-based, purely data-driven, and hybrid approaches to parameter estimation and filtering, where the latter tries to combine the benefits of model-based signal processing and data-driven deep learning to overcome their individual deficiencies. We illustrate the underlying design principles with examples from noise reduction, source separation, and dereverberation.

Keywords

Cite

@article{arxiv.2501.07215,
  title  = {Microphone Array Signal Processing and Deep Learning for Speech Enhancement},
  author = {Reinhold Haeb-Umbach and Tomohiro Nakatani and Marc Delcroix and Christoph Boeddeker and Tsubasa Ochiai},
  journal= {arXiv preprint arXiv:2501.07215},
  year   = {2025}
}

Comments

Accepted for IEEE Signal Processing Magazine

R2 v1 2026-06-28T21:04:28.839Z