TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

SotA/Audio/Speech Enhancement/Deep Noise Suppression (DNS) Challenge

Speech Enhancement on Deep Noise Suppression (DNS) Challenge

Metric: SI-SDR-WB (higher is better)

LeaderboardDataset
Loading chart...

Results

Submit a result
#Model↕SI-SDR-WB▼Extra DataPaperDate↕Code
1TF-Locoformer (M)23.3NoTF-Locoformer: Transformer with Local Modeling b...2024-08-06Code
2ZipEnhancer (M)22.22No---
3BSRNN-S + MRSD21.4NoHigh Fidelity Speech Enhancement with Band-split...2022-12-01Code
4BSRNN-S + MGD21.4NoHigh Fidelity Speech Enhancement with Band-split...2022-12-01Code
5BSRNN-S21.3NoHigh Fidelity Speech Enhancement with Band-split...2022-12-01Code
6MambAttention21.234NoMambAttention: Mamba with Multi-Head Attention f...2025-07-01Code
7ZipEnhancer (S)21.15No---
8BSRNN-16k21.1NoHigh Fidelity Speech Enhancement with Band-split...2022-12-01Code
9MP-SENet21.03NoExplicit Estimation of Magnitude and Phase Spect...2023-08-17Code
10xLSTM-SENet20.854NoMambAttention: Mamba with Multi-Head Attention f...2025-07-01Code
11MFNET20.31NoA Mask Free Neural Network for Monaural Speech E...2023-06-07Code
12Sudo rm -rf (U=32)19.7NoRemixIT: Continual self-training of speech enhan...2022-02-17Code
13Sudo rm-rf (U=8)18.6NoContinual self-training with bootstrapped remixi...2021-10-19Code
14RemixIT (w Sudo U=32)18YesContinual self-training with bootstrapped remixi...2021-10-19Code
15FullSubNet17.29NoFullSubNet: A Full-Band and Sub-Band Fusion Mode...2020-10-29Code
16FullSubNet+16.81NoFullSubNet+: Channel Attention FullSubNet with C...2022-03-23Code
17DTLN16.34No--Code
18Non-Real-Time MultiScale+16.22NoPhase-aware Single-stage Speech Denoising and De...2020-06-01Code
19RemixIT (w Sudo U=32)16YesRemixIT: Continual self-training of speech enhan...2022-02-17Code
20Noisy9.1No---