Robust Online Overdetermined Independent Vector Analysis Based on Bilinear Decomposition

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Kang, Wang, Xianrui, Yang, Yichen, Brendel, Andreas, Huang, Gongping, Koldovský, Zbyněk, Chen, Jingdong, Benesty, Jacob, Makino, Shoji
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909994296803328
author Chen, Kang
Wang, Xianrui
Yang, Yichen
Brendel, Andreas
Huang, Gongping
Koldovský, Zbyněk
Chen, Jingdong
Benesty, Jacob
Makino, Shoji
author_facet Chen, Kang
Wang, Xianrui
Yang, Yichen
Brendel, Andreas
Huang, Gongping
Koldovský, Zbyněk
Chen, Jingdong
Benesty, Jacob
Makino, Shoji
contents Online blind source separation is essential for both speech communication and human-machine interaction. Among existing approaches, overdetermined independent vector analysis (OverIVA) delivers strong performance by exploiting the statistical independence of source signals and the orthogonality between source and noise subspaces. However, when applied to large microphone arrays, the number of parameters grows rapidly, which can degrade online estimation accuracy. To overcome this challenge, we propose decomposing each long separation filter into a bilinear form of two shorter filters, thereby reducing the number of parameters. Because the two filters are closely coupled, we design an alternating iterative projection algorithm to update them in turn. Simulation results show that, with far fewer parameters, the proposed method achieves improved performance and robustness.
format Preprint
id arxiv_https___arxiv_org_abs_2601_12485
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Robust Online Overdetermined Independent Vector Analysis Based on Bilinear Decomposition
Chen, Kang
Wang, Xianrui
Yang, Yichen
Brendel, Andreas
Huang, Gongping
Koldovský, Zbyněk
Chen, Jingdong
Benesty, Jacob
Makino, Shoji
Audio and Speech Processing
Sound
Online blind source separation is essential for both speech communication and human-machine interaction. Among existing approaches, overdetermined independent vector analysis (OverIVA) delivers strong performance by exploiting the statistical independence of source signals and the orthogonality between source and noise subspaces. However, when applied to large microphone arrays, the number of parameters grows rapidly, which can degrade online estimation accuracy. To overcome this challenge, we propose decomposing each long separation filter into a bilinear form of two shorter filters, thereby reducing the number of parameters. Because the two filters are closely coupled, we design an alternating iterative projection algorithm to update them in turn. Simulation results show that, with far fewer parameters, the proposed method achieves improved performance and robustness.
title Robust Online Overdetermined Independent Vector Analysis Based on Bilinear Decomposition
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2601.12485