CleanUMamba

A Compact Mamba Network for Speech Denoising using Channel Pruning

Conference Paper (2025)
Author(s)

Sjoerd Groot (Student TU Delft)

Qinyu Chen (Universiteit Leiden)

Jan Van Gemert (TU Delft - Pattern Recognition and Bioinformatics)

C. Gao (TU Delft - Electronics)

Research Group
Pattern Recognition and Bioinformatics
DOI related publication
https://doi.org/10.1109/ISCAS56072.2025.11043389
More Info
expand_more
Publication Year
2025
Language
English
Research Group
Pattern Recognition and Bioinformatics
Bibliographical Note
Green Open Access added to TU Delft Institutional Repository as part of the Taverne amendment. More information about this copyright law amendment can be found at https://www.openaccess.nl. Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.@en
ISBN (print)
979-8-3503-5684-7
ISBN (electronic)
979-8-3503-5683-0
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

This paper presents CleanUMamba, a time-domain neural network architecture designed for real-time causal audio denoising directly applied to raw waveforms. CleanUMamba leverages a U-Net encoder-decoder structure, incorporating the Mamba state-space model in the bottleneck layer. By replacing conventional self-attention and LSTM mechanisms with Mamba, our architecture offers superior denoising performance while maintaining a constant memory footprint, enabling streaming operation. To enhance efficiency, we applied structured channel pruning, achieving an 8X reduction in model size without compromising audio quality. Our model demonstrates strong results in the Interspeech 2020 Deep Noise Suppression challenge. Specifically, CleanUMamba achieves a PESQ score of 2.42 and STOI of 95.1% with only 442K parameters and 468M MACs, matching or outperforming larger models in real-time performance. Code will be available at: https://github.com/lab-emi/CleanUMamba

Files

License info not available
warning

File under embargo until 27-12-2025