# LiSenNet: Lightweight Sub-band and Dual-Path Modeling for Real-Time Speech Enhancement

arXiv: https://arxiv.org/abs/2409.13285 · HTML: https://mlsanity.com/paper/2409.13285/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Sep 2024.

## Built on by (most cited first, 10 of 10)

- [A Composite Predictive-Generative Approach to Monaural Universal Speech Enhancement](https://mlsanity.com/paper/2505.24576/) (2025)
- [CleanMel: Mel-Spectrogram Enhancement for Improving Both Speech Quality and ASR](https://mlsanity.com/paper/2502.20040/) (2025)
- [MambAttention: Mamba with Multi-Head Attention for Generalizable Single-Channel Speech Enhancement](https://mlsanity.com/paper/2507.00966/) (2025)
- [Quaternion Self-Attention with Shared Scores](https://mlsanity.com/paper/2605.24920/) (2026)
- [UL-UNAS: Ultra-Lightweight U-Nets for Real-Time Speech Enhancement via Network Architecture Search](https://mlsanity.com/paper/2503.00340/) (2025)
- [Single Microphone Own Voice Detection based on Simulated Transfer Functions for Hearing Aids](https://mlsanity.com/paper/2603.02724/) (2026)
- [BioSEN: A Bio-acoustic Signal Enhancement Network for Animal Vocalizations](https://mlsanity.com/paper/2605.12534/) (2026)
- [QC-GAN: A Parameter-Efficient Quaternion Conformer GAN for High-Fidelity Speech Enhancement](https://mlsanity.com/paper/2606.18611/) (2026)
- [RT-SEMamba: Real-Time Speech Enhancement Mamba via Progressive Knowledge Distillation](https://mlsanity.com/paper/2608.12099/) (2026)
- [Beyond Model Size: Redesigning LiSenNet for embedded speech enhancement](https://mlsanity.com/paper/2609.29866/) (2026)

## Abstract (arXiv)

Speech enhancement (SE) aims to extract the clean waveform from noise-contaminated measurements to improve the speech quality and intelligibility. Although learning-based methods can perform much better than traditional counterparts, the large computational complexity and model size heavily limit the deployment on latency-sensitive and low-resource edge devices. In this work, we propose a lightweight SE network (LiSenNet) for real-time applications. We design sub-band downsampling and upsampling blocks and a dual-path recurrent module to capture band-aware features and time-frequency patterns, respectively. A noise detector is developed to detect noisy regions in order to perform SE adaptively and save computational costs. Compared to recent higher-resource-dependent baseline models, the proposed LiSenNet can achieve a competitive performance with only 37k parameters (half of the state-of-the-art model) and 56M multiply-accumulate (MAC) operations per second.