Controlling the Parameterized Multi-channel Wiener Filter using a tiny neural network

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Grinstein, Eric, Pandey, Ashutosh, Li, Cole, Srinivas, Shanmukha, Azcarreta, Juan, Donley, Jacob, Lee, Sanha, Aroudi, Ali, Bilen, Cagdas
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908455606943744
author Grinstein, Eric
Pandey, Ashutosh
Li, Cole
Srinivas, Shanmukha
Azcarreta, Juan
Donley, Jacob
Lee, Sanha
Aroudi, Ali
Bilen, Cagdas
author_facet Grinstein, Eric
Pandey, Ashutosh
Li, Cole
Srinivas, Shanmukha
Azcarreta, Juan
Donley, Jacob
Lee, Sanha
Aroudi, Ali
Bilen, Cagdas
contents Noise suppression and speech distortion are two important aspects to be balanced when designing multi-channel Speech Enhancement (SE) algorithms. Although neural network models have achieved state-of-the-art noise suppression, their non-linear operations often introduce high speech distortion. Conversely, classical signal processing algorithms such as the Parameterized Multi-channel Wiener Filter ( PMWF) beamformer offer explicit mechanisms for controlling the suppression/distortion trade-off. In this work, we present NeuralPMWF, a system where the PMWF is entirely controlled using a low-latency, low-compute neural network, resulting in a low-complexity system offering high noise reduction and low speech distortion. Experimental results show that our proposed approach results in significantly better perceptual and objective speech enhancement in comparison to several competitive baselines using similar computational resources.
format Preprint
id arxiv_https___arxiv_org_abs_2507_13863
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Controlling the Parameterized Multi-channel Wiener Filter using a tiny neural network
Grinstein, Eric
Pandey, Ashutosh
Li, Cole
Srinivas, Shanmukha
Azcarreta, Juan
Donley, Jacob
Lee, Sanha
Aroudi, Ali
Bilen, Cagdas
Sound
Audio and Speech Processing
Noise suppression and speech distortion are two important aspects to be balanced when designing multi-channel Speech Enhancement (SE) algorithms. Although neural network models have achieved state-of-the-art noise suppression, their non-linear operations often introduce high speech distortion. Conversely, classical signal processing algorithms such as the Parameterized Multi-channel Wiener Filter ( PMWF) beamformer offer explicit mechanisms for controlling the suppression/distortion trade-off. In this work, we present NeuralPMWF, a system where the PMWF is entirely controlled using a low-latency, low-compute neural network, resulting in a low-complexity system offering high noise reduction and low speech distortion. Experimental results show that our proposed approach results in significantly better perceptual and objective speech enhancement in comparison to several competitive baselines using similar computational resources.
title Controlling the Parameterized Multi-channel Wiener Filter using a tiny neural network
topic Sound
Audio and Speech Processing
url https://arxiv.org/abs/2507.13863