SOTAVerified

Speech Enhancement

Speech Enhancement is a signal processing task that involves improving the quality of speech signals captured under noisy or degraded conditions. The goal of speech enhancement is to make speech signals clearer, more intelligible, and more pleasant to listen to, which can be used for various applications such as voice recognition, teleconferencing, and hearing aids. A representative Github project with online demo : ClearerVoice-Studio.

( Image credit: A Fully Convolutional Neural Network For Speech Enhancement )

Papers

Showing 125 of 982 papers

TitleStatusHype
Autoregressive Speech Enhancement via Acoustic Tokens0
P.808 Multilingual Speech Enhancement Testing: Approach and Results of URGENT 2025 Challenge0
Speech Quality Assessment Model Based on Mixture of Experts: System-Level Performance Enhancement and Utterance-Level Challenge Analysis0
Robust One-step Speech Enhancement via Consistency DistillationCode1
MambAttention: Mamba with Multi-Head Attention for Generalizable Single-Channel Speech EnhancementCode2
Frequency-Weighted Training Losses for Phoneme-Level DNN-based Speech Enhancement0
EDNet: A Distortion-Agnostic Speech Enhancement Framework with Gating Mamba Mechanism and Phase Shift-Invariant Training0
A Comparative Evaluation of Deep Learning Models for Speech Enhancement in Real-World Noisy Environments0
Efficient Speech Enhancement via Embeddings from Pre-trained Generative AudioencodersCode2
Exploring Length Generalization For Transformer-based Speech Enhancement0
French Listening Tests for the Assessment of Intelligibility, Quality, and Identity of Body-Conducted Speech Enhancement0
Diffusion Buffer: Online Diffusion-based Speech Enhancement with Sub-Second Latency0
Lessons Learned from the URGENT 2024 Speech Enhancement ChallengeCode0
A Two-Stage Hierarchical Deep Filtering Framework for Real-Time Speech Enhancement0
A Composite Predictive-Generative Approach to Monaural Universal Speech Enhancement0
Boosting Domain Incremental Learning: Selecting the Optimal Parameters is All You NeedCode0
DeepFilterGAN: A Full-band Real-time Speech Enhancement System with GAN-based Stochastic Regeneration0
Interspeech 2025 URGENT Speech Enhancement Challenge0
ARiSE: Auto-Regressive Multi-Channel Speech Enhancement0
Model as Loss: A Self-Consistent Training Paradigm0
Study of Lightweight Transformer Architectures for Single-Channel Speech Enhancement0
A Lightweight Hybrid Dual Channel Speech Enhancement System under Low-SNR ConditionsCode2
FlowSE: Efficient and High-Quality Speech Enhancement via Flow MatchingCode2
Stack Less, Repeat More: A Block Reusing Approach for Progressive Speech Enhancement0
Mel-McNet: A Mel-Scale Framework for Online Multichannel Speech EnhancementCode1
Show:102550
← PrevPage 1 of 40Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ROSE-CD(PESQ)PESQ (wb)3.99Unverified
2PESQetarianPESQ (wb)3.82Unverified
3Mamba-SEUNet L (+PCS)PESQ (wb)3.73Unverified
4Schrödinger bridge (PESQ loss)PESQ (wb)3.7Unverified
5SEMamba (+PCS)PESQ (wb)3.69Unverified
6ZipEnhancer (S, \lamba_6 = 0)PESQ (wb)3.63Unverified
7PrimeK-NetPESQ (wb)3.61Unverified
8ZipEnhancer (S, \lamba_6 = 0.2)PESQ (wb)3.61Unverified
9MP-SENetPESQ (wb)3.6Unverified
10PCS_CS_WAVLMPESQ (wb)3.54Unverified
#ModelMetricClaimedVerifiedStatus
1BSRNN-S + MGDSI-SDR-WB21.4Unverified
2DTLNSI-SDR-WB16.34Unverified
3Non-Real-Time MultiScale+SI-SDR-WB16.22Unverified
4ZipEnhancer (M)PESQ-WB3.81Unverified
5TF-Locoformer (M)PESQ-WB3.72Unverified
6ZipEnhancer (S)PESQ-WB3.69Unverified
7MambAttentionPESQ-WB3.67Unverified
8MP-SENetPESQ-WB3.62Unverified
9xLSTM-SENetPESQ-WB3.59Unverified
10BSRNN-S + MRSDPESQ-WB3.53Unverified
#ModelMetricClaimedVerifiedStatus
1Inter-Channel Conv-TasNetSDR19.67Unverified
2CA Dense U-Net (Complex)SDR18.64Unverified
3Dense U-Net (Complex)SDR18.4Unverified
4Dense U-Net (Real)SDR16.86Unverified
5U-Net (Real)SDR15.97Unverified
6Noisy/unprocessedSDR6.5Unverified
#ModelMetricClaimedVerifiedStatus
1Schrödinger Bridge (PESQ loss)PESQ-WB3.09Unverified
2SGMSE+PESQ-WB2.5Unverified
3Demucs v4PESQ-WB2.37Unverified
4Schrödinger BridgePESQ-WB2.33Unverified
5Conv-TasNetPESQ-WB2.31Unverified
6CDiffuSEPESQ-WB1.6Unverified
#ModelMetricClaimedVerifiedStatus
1ReVISE (ch2)Audio Quality MOS4.19Unverified
2ReVISE (bf)Audio Quality MOS4.11Unverified
3Demucs (ch2)Audio Quality MOS2.95Unverified
4Demucs (bf)Audio Quality MOS2.39Unverified
5MaxDI (Baseline)PESQ1.17Unverified
6DAJA (MVDR,HMA,1000) (Overlapped Speech)SDR-4.76Unverified
#ModelMetricClaimedVerifiedStatus
1ZipEnhancer (M)PESQ-NB4.08Unverified
2DCCRN-MCPESQ-NB3.21Unverified
3DCCRN-MPESQ-NB3.15Unverified
4DCCRNPESQ-NB3.04Unverified
5RNN-ModulationPESQ-WB2.75Unverified
#ModelMetricClaimedVerifiedStatus
1MambAttentionESTOI0.8Unverified
2SEMambaESTOI0.8Unverified
3xLSTM-SENetESTOI0.8Unverified
4MP-SENetESTOI0.79Unverified
#ModelMetricClaimedVerifiedStatus
1SepFormerPESQ2.84Unverified
2DTLNPESQ2.23Unverified
3UnprocessedPESQ1.83Unverified
4Non-Real-Time MultiScale+PESQ1.52Unverified
#ModelMetricClaimedVerifiedStatus
1DCUNet-MCPESQ-NB3.44Unverified
2DCCRN-MPESQ-NB3.28Unverified
3DCUNetPESQ-NB3.25Unverified
#ModelMetricClaimedVerifiedStatus
1CleanMel-L-mapDNSMOS3.82Unverified
2SpatialNetDNSMOS BAK3.43Unverified
#ModelMetricClaimedVerifiedStatus
1rose_cd(PESQ )PESQ3.99Unverified
2ROSE-CDPESQ3.49Unverified
#ModelMetricClaimedVerifiedStatus
1Wave-U-NetCBAK3.24Unverified
#ModelMetricClaimedVerifiedStatus
1Audio-Visual concat-refPESQ2.7Unverified
#ModelMetricClaimedVerifiedStatus
1SE-MelGANAudio Quality MOS3.1Unverified
#ModelMetricClaimedVerifiedStatus
1DeFT-ANPESQ3.01Unverified
#ModelMetricClaimedVerifiedStatus
1Audio-Visual concat-refPESQ3.03Unverified
#ModelMetricClaimedVerifiedStatus
1SepFormerPESQ3.07Unverified