On Joint Dereverberation and Single Moving Source Separation with Online Source Steering

To solve the Blind Source Separation (BSS) problems in time-variant and highly reverberant environments, an online joint optimization algorithm for Weighted Prediction Error (WPE)-based dereverberation and Independent Vector Analysis (IVA)-based source separation (online WPE-IVA) has been reported. On the other hand, Online Source Steering (OSS) was recently introduced for filter updating in online IVA with low computational cost for a target-source tracking task where only one speaker is moving. Inspired by this work, we integrate OSS into online WPE-IVA, aiming to reduce the computational cost of updating separation filters. Experimental results show that our proposed method reduces computational costs while achieving comparable separation performance to conventional online WPE-IVA in target-source tracking tasks in highly reverberant environments.

Authors

Institutions

Publication Details

Published
2024-12-03
DOI
https://doi.org/10.1109/apsipaasc63619.2025.10848715
Primary Topic
Speech and Audio Processing
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

On Joint Dereverberation and Single Moving Source Separation with Online Source Steering

Yiting Zhang, Shoji Makino, Kaien Mo, T. Ueda et al.
Speech and Audio Processing
article

On Joint Dereverberation and Single Moving Source Separation with Online Source Steering

Yiting Zhang, Shoji Makino, Kaien Mo, T. Ueda, Yichen Yang
article en

Abstract

To solve the Blind Source Separation (BSS) problems in time-variant and highly reverberant environments, an online joint optimization algorithm for Weighted Prediction Error (WPE)-based dereverberation and Independent Vector Analysis (IVA)-based source separation (online WPE-IVA) has been reported. On the other hand, Online Source Steering (OSS) was recently introduced for filter updating in online IVA with low computational cost for a target-source tracking task where only one speaker is moving. Inspired by this work, we integrate OSS into online WPE-IVA, aiming to reduce the computational cost of updating separation filters. Experimental results show that our proposed method reduces computational costs while achieving comparable separation performance to conventional online WPE-IVA in target-source tracking tasks in highly reverberant environments.

Waseda University (JP)
Openalex Percentile: Top 71%
Speech and Audio Processing
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.