Saved in:
Bibliographic Details
Main Authors: Eisenberg, Aviad, Gannot, Sharon, Chazan, Shlomo E.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.20165
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914217007775744
author Eisenberg, Aviad
Gannot, Sharon
Chazan, Shlomo E.
author_facet Eisenberg, Aviad
Gannot, Sharon
Chazan, Shlomo E.
contents This paper presents a robust multi-channel speaker extraction algorithm designed to handle inaccuracies in reference information. While existing approaches often rely solely on either spatial or spectral cues to identify the target speaker, our method integrates both sources of information to enhance robustness. A key aspect of our approach is its emphasis on stability, ensuring reliable performance even when one of the features is degraded or misleading. Given a noisy mixture and two potentially unreliable cues, a dedicated network is trained to dynamically balance their contributions-or disregard the less informative one when necessary. We evaluate the system under challenging conditions by simulating inference-time errors using a simple direction of arrival (DOA) estimator and a noisy spectral enrollment process. Experimental results demonstrate that the proposed model successfully extracts the desired speaker even in the presence of substantial reference inaccuracies.
format Preprint
id arxiv_https___arxiv_org_abs_2512_20165
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Spectral or spatial? Leveraging both for speaker extraction in challenging data conditions
Eisenberg, Aviad
Gannot, Sharon
Chazan, Shlomo E.
Sound
Audio and Speech Processing
This paper presents a robust multi-channel speaker extraction algorithm designed to handle inaccuracies in reference information. While existing approaches often rely solely on either spatial or spectral cues to identify the target speaker, our method integrates both sources of information to enhance robustness. A key aspect of our approach is its emphasis on stability, ensuring reliable performance even when one of the features is degraded or misleading. Given a noisy mixture and two potentially unreliable cues, a dedicated network is trained to dynamically balance their contributions-or disregard the less informative one when necessary. We evaluate the system under challenging conditions by simulating inference-time errors using a simple direction of arrival (DOA) estimator and a noisy spectral enrollment process. Experimental results demonstrate that the proposed model successfully extracts the desired speaker even in the presence of substantial reference inaccuracies.
title Spectral or spatial? Leveraging both for speaker extraction in challenging data conditions
topic Sound
Audio and Speech Processing
url https://arxiv.org/abs/2512.20165