Joint Dereverberation and Separation with Iterative Source Steering

by   Taishi Nakashima, et al.

We propose a new algorithm for joint dereverberation and blind source separation (DR-BSS). Our work builds upon the IRLMA-T framework that applies a unified filter combining dereverberation and separation. One drawback of this framework is that it requires several matrix inversions, an operation inherently costly and with potential stability issues. We leverage the recently introduced iterative source steering (ISS) updates to propose two algorithms mitigating this issue. Albeit derived from first principles, the first algorithm turns out to be a natural combination of weighted prediction error (WPE) dereverberation and ISS-based BSS, applied alternatingly. In this case, we manage to reduce the number of matrix inversion to only one per iteration and source. The second algorithm updates the ILRMA-T matrix using only sequential ISS updates requiring no matrix inversion at all. Its implementation is straightforward and memory efficient. Numerical experiments demonstrate that both methods achieve the same final performance as ILRMA-T in terms of several relevant objective metrics. In the important case of two sources, the number of iterations required is also similar.



There are no comments yet.


page 1

page 2

page 3

page 4


A Joint Diagonalization Based Efficient Approach to Underdetermined Blind Audio Source Separation Using the Multichannel Wiener Filter

This paper presents a computationally efficient approach to blind source...

Efficient Full-Rank Spatial Covariance Estimation Using Independent Low-Rank Matrix Analysis for Blind Source Separation

In this paper, we propose a new algorithm that efficiently separates a d...

Regularized Fast Multichannel Nonnegative Matrix Factorization with ILRMA-based Prior Distribution of Joint-Diagonalization Process

In this paper, we address a convolutive blind source separation (BSS) pr...

SDR – Medium Rare with Fast Computations

We revisit the widely used bss eval metrics for source separation with a...

Blind Source Separation Algorithms Using Hyperbolic and Givens Rotations for High-Order QAM Constellations

This paper addresses the problem of blind demixing of instantaneous mixt...

Towards the Application of Linear Programming Methods For Multi-Camera Pose Estimation

We presented a separation based optimization algorithm which, rather tha...

FastFCA-AS: Joint Diagonalization Based Acceleration of Full-Rank Spatial Covariance Analysis for Separating Any Number of Sources

Here we propose FastFCA-AS, an accelerated algorithm for Full-rank spati...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.