{"url":"/sota/speech-enhancement-on-demand","task":{"name":"Speech Enhancement","url":"/task/speech-enhancement","note":null},"dataset":{"name":"VoiceBank + DEMAND","url":"/dataset/demand"},"category":"Audio","categories":["Audio","Speech"],"category_note":null,"description":"**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](https://github.com/modelscope/ClearerVoice-Studio).\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [A Fully Convolutional Neural Network For Speech Enhancement](https://arxiv.org/pdf/1609.07132v1.pdf) )</span>","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["PESQ (wb)","CBAK","COVL","CSIG","STOI","ESTOI","SSNR","SI-SDR","Para. (M)"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"PESQ (wb)":null,"CBAK":null,"COVL":null,"CSIG":null,"STOI":null,"ESTOI":null,"SSNR":null,"SI-SDR":null,"Para. (M)":null}},"counts":{"rows":42,"rows_with_code":31,"rows_with_paper_page":39,"rows_dated":39,"rows_using_additional_data":1},"rows":[{"rank_in_archive_order":1,"model":"ROSE-CD(PESQ)","metrics":{"CBAK":"3.37","COVL":"4.30","CSIG":"4.63","ESTOI":"0.83","PESQ (wb)":"3.99","Para. (M)":"65","SI-SDR":"0.40","SSNR":"0.927","STOI":"92.6"},"uses_additional_data":false,"paper_date":"2025-07-08","paper":"/paper/robust-one-step-speech-enhancement-via-1","paper_url":"https://arxiv.org/abs/2507.05688v1","paper_title":"Robust One-step Speech Enhancement via Consistency Distillation","code":"https://github.com/LiangXu123/Robust-One-step-Speech-Enhancement-via-Consistency-Distillation-ROSE-CD-","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"PESQetarian","metrics":{"CBAK":"2.49","COVL":"3.5","CSIG":"3.63","ESTOI":"0.84","PESQ (wb)":"3.82","Para. (M)":"30","SI-SDR":"-19.8","SSNR":"-2.72","STOI":"0.92"},"uses_additional_data":false,"paper_date":"2024-06-05","paper":"/paper/the-pesqetarian-on-the-relevance-of-goodhart","paper_url":"https://arxiv.org/abs/2406.03460v1","paper_title":"The PESQetarian: On the Relevance of Goodhart's Law for Speech Enhancement","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":3,"model":"Mamba-SEUNet L (+PCS)","metrics":{"CBAK":"3.67","COVL":"4.40","CSIG":"4.82","PESQ (wb)":"3.73","Para. (M)":"6.28","STOI":"96"},"uses_additional_data":false,"paper_date":"2024-12-21","paper":"/paper/mamba-seunet-mamba-unet-for-monaural-speech","paper_url":"https://arxiv.org/abs/2412.16626v2","paper_title":"Mamba-SEUNet: Mamba UNet for Monaural Speech Enhancement","code":"https://github.com/MyParadise21/Mamba-SEUNet","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"Schrödinger bridge (PESQ loss)","metrics":{"PESQ (wb)":"3.70"},"uses_additional_data":false,"paper_date":"2024-09-16","paper":"/paper/investigating-training-objectives-for","paper_url":"https://arxiv.org/abs/2409.10753v2","paper_title":"Investigating Training Objectives for Generative Speech Enhancement","code":"https://github.com/sp-uhh/sgmse","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"SEMamba (+PCS)","metrics":{"CBAK":"3.63","COVL":"4.37","CSIG":"4.79","PESQ (wb)":"3.69","Para. (M)":"2.25","STOI":"96"},"uses_additional_data":false,"paper_date":"2024-05-10","paper":"/paper/an-investigation-of-incorporating-mamba-for","paper_url":"https://arxiv.org/abs/2405.06573v1","paper_title":"An Investigation of Incorporating Mamba for Speech Enhancement","code":"https://github.com/roychao19477/semamba","n_code_links":1,"syntology":{"n_ran":13,"n_unverified":1,"n_samples":14,"n_pointer_only_licence":14}},{"rank_in_archive_order":6,"model":"ZipEnhancer (S, \\lamba_6 = 0)","metrics":{"CBAK":"3.87","COVL":"4.36","CSIG":"4.81","PESQ (wb)":"3.63","Para. (M)":"2.04","SI-SDR":"19.09","SSNR":"8.33","STOI":"96.19"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"PrimeK-Net","metrics":{"CBAK":"3.98","COVL":"4.35","CSIG":"4.81","PESQ (wb)":"3.61","Para. (M)":"1.41","STOI":"96"},"uses_additional_data":false,"paper_date":"2025-02-27","paper":"/paper/primek-net-multi-scale-spectral-learning-via","paper_url":"https://arxiv.org/abs/2502.19906v1","paper_title":"PrimeK-Net: Multi-scale Spectral Learning via Group Prime-Kernel Convolutional Neural Networks for Single Channel Speech Enhancement","code":"https://github.com/huaidanquede/PrimeK-Net","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"ZipEnhancer (S, \\lamba_6 = 0.2)","metrics":{"CBAK":"3.97","COVL":"4.35","CSIG":"4.81","PESQ (wb)":"3.61","Para. (M)":"2.04","SI-SDR":"19.96","SSNR":"10.01","STOI":"96.22"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"MP-SENet","metrics":{"CBAK":"3.99","COVL":"4.34","CSIG":"4.81","PESQ (wb)":"3.60","Para. (M)":"2.26","STOI":"0.96"},"uses_additional_data":false,"paper_date":"2023-08-17","paper":"/paper/explicit-estimation-of-magnitude-and-phase","paper_url":"https://arxiv.org/abs/2308.08926v2","paper_title":"Explicit Estimation of Magnitude and Phase Spectra in Parallel for High-Quality Speech Enhancement","code":"https://github.com/yxlu-0102/MP-SENet","n_code_links":1,"syntology":{"n_ran":13,"n_unverified":1,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"PCS_CS_WAVLM","metrics":{"CBAK":"3.49","COVL":"4.20","CSIG":"4.75","PESQ (wb)":"3.54","STOI":"0.96"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":11,"model":"xLSTM-SENet2","metrics":{"CBAK":"3.98","COVL":"4.27","CSIG":"4.78","PESQ (wb)":"3.53","Para. (M)":"2.27","STOI":"0.96"},"uses_additional_data":false,"paper_date":"2025-01-10","paper":"/paper/xlstm-senet-xlstm-for-single-channel-speech","paper_url":"https://arxiv.org/abs/2501.06146v2","paper_title":"xLSTM-SENet: xLSTM for Single-Channel Speech Enhancement","code":"https://github.com/nikolaikyhne/xlstm-senet","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"SCP-CMGAN","metrics":{"CBAK":"3.97","COVL":"4.25","CSIG":"4.75","PESQ (wb)":"3.52","SSNR":"10.82","STOI":"96"},"uses_additional_data":false,"paper_date":"2022-10-26","paper":"/paper/scp-gan-self-correcting-discriminator","paper_url":"https://arxiv.org/abs/2210.14474v1","paper_title":"SCP-GAN: Self-Correcting Discriminator Optimization for Training Consistency Preserving Metric GAN on Speech Enhancement Tasks","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"ROSE-CD","metrics":{"CBAK":"3.33","COVL":"4.04","CSIG":"4.523","ESTOI":"0.87","PESQ (wb)":"3.49","Para. (M)":"65","SI-SDR":"17.80","SSNR":"3.34","STOI":"94.73"},"uses_additional_data":false,"paper_date":"2025-07-08","paper":"/paper/robust-one-step-speech-enhancement-via-1","paper_url":"https://arxiv.org/abs/2507.05688v1","paper_title":"Robust One-step Speech Enhancement via Consistency Distillation","code":"https://github.com/LiangXu123/Robust-One-step-Speech-Enhancement-via-Consistency-Distillation-ROSE-CD-","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"D2Former","metrics":{"PESQ (wb)":"3.43","Para. (M)":"0.86"},"uses_additional_data":false,"paper_date":"2021-02-03","paper":"/paper/monaural-speech-enhancement-with-complex","paper_url":"https://arxiv.org/abs/2102.01993v2","paper_title":"Monaural Speech Enhancement with Complex Convolutional Block Attention Module and Joint Time Frequency Losses","code":"https://github.com/modelscope/ClearerVoice-Studio","n_code_links":2,"syntology":null},{"rank_in_archive_order":15,"model":"CMGAN","metrics":{"CBAK":"3.94","COVL":"4.12","CSIG":"4.63","PESQ (wb)":"3.41","SSNR":"11.1","STOI":"96"},"uses_additional_data":false,"paper_date":"2022-09-22","paper":"/paper/cmgan-conformer-based-metric-gan-for-monaural","paper_url":"https://arxiv.org/abs/2209.11112v3","paper_title":"CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancement","code":"https://github.com/ruizhecao96/cmgan","n_code_links":2,"syntology":null},{"rank_in_archive_order":16,"model":"PCS","metrics":{"COVL":"3.92","CSIG":"4.43","PESQ (wb)":"3.35","STOI":"95"},"uses_additional_data":false,"paper_date":"2022-03-31","paper":"/paper/perceptual-contrast-stretching-on-target","paper_url":"https://arxiv.org/abs/2203.17152v4","paper_title":"Perceptual Contrast Stretching on Target Feature for Speech Enhancement","code":"https://github.com/roychao19477/pcs","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"D²Net","metrics":{"CBAK":"3.18","COVL":"3.92","CSIG":"4.63","PESQ (wb)":"3.27","STOI":"96"},"uses_additional_data":false,"paper_date":"2022-11-21","paper":"/paper/d2net-a-denoising-and-dereverberation-network","paper_url":"https://ieeexplore.ieee.org/abstract/document/9979863","paper_title":"D²Net: A Denoising and Dereverberation Network Based on Two-branch Encoder and Dual-path Transformer","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":18,"model":"aTENNuate","metrics":{"CBAK":"2.85","COVL":"3.96","CSIG":"4.57","PESQ (wb)":"3.27","SI-SDR":"15.04"},"uses_additional_data":false,"paper_date":"2024-09-05","paper":"/paper/raw-speech-enhancement-with-deep-state-space","paper_url":"https://arxiv.org/abs/2409.03377v4","paper_title":"aTENNuate: Optimized Real-time Speech Enhancement with Deep SSMs on Raw Audio","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":19,"model":"Centaurus (0.51M)","metrics":{"PESQ (wb)":"3.25"},"uses_additional_data":false,"paper_date":"2025-01-22","paper":"/paper/let-ssms-be-convnets-state-space-modeling","paper_url":"https://arxiv.org/abs/2501.13230v1","paper_title":"Let SSMs be ConvNets: State-space Modeling with Optimal Tensor Contractions","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":20,"model":"MetricGAN-OKD","metrics":{"CBAK":"3.07","COVL":"3.73","CSIG":"4.23","PESQ (wb)":"3.24","Para. (M)":"1.89"},"uses_additional_data":false,"paper_date":"2023-07-24","paper":"/paper/metricgan-okd-multi-metric-optimization-of","paper_url":"https://proceedings.mlr.press/v202/shin23b.html","paper_title":"MetricGAN-OKD: Multi-Metric Optimization of MetricGAN via Online Knowledge Distillation for Speech Enhancement","code":"https://github.com/wooseok-shin/MetricGAN-OKD","n_code_links":1,"syntology":null},{"rank_in_archive_order":21,"model":"MANNER","metrics":{"CBAK":"3.65","COVL":"3.91","CSIG":"4.53","PESQ (wb)":"3.21","STOI":"95"},"uses_additional_data":false,"paper_date":"2022-03-04","paper":"/paper/manner-multi-view-attention-network-for-noise","paper_url":"https://arxiv.org/abs/2203.02181v1","paper_title":"MANNER: Multi-view Attention Network for Noise Erasure","code":"https://github.com/winddori2002/MANNER","n_code_links":1,"syntology":null},{"rank_in_archive_order":22,"model":"BSSE-SE","metrics":{"CBAK":"3.58","COVL":"3.88","CSIG":"4.52","PESQ (wb)":"3.20","STOI":"95.7"},"uses_additional_data":false,"paper_date":"2022-04-07","paper":"/paper/boosting-self-supervised-embeddings-for","paper_url":"https://arxiv.org/abs/2204.03339v2","paper_title":"Boosting Self-Supervised Embeddings for Speech Enhancement","code":"https://github.com/khhungg/BSSE-SE","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":5,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":23,"model":"DeepFilterNet3","metrics":{"CBAK":"3.61","COVL":"3.77","CSIG":"4.34","PESQ (wb)":"3.17","STOI":"0.944"},"uses_additional_data":false,"paper_date":"2023-05-14","paper":"/paper/deepfilternet-perceptually-motivated-real","paper_url":"https://arxiv.org/abs/2305.08227v1","paper_title":"DeepFilterNet: Perceptually Motivated Real-Time Speech Enhancement","code":"https://github.com/rikorose/deepfilternet","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"PERL-AE","metrics":{"CBAK":"3.53","COVL":"3.83","CSIG":"4.43","PESQ (wb)":"3.17"},"uses_additional_data":true,"paper_date":"2020-10-22","paper":"/paper/perceptual-loss-based-speech-denoising-with","paper_url":"http://arxiv.org/abs/2010.11860v1","paper_title":"Perceptual Loss based Speech Denoising with an ensemble of Audio Pattern Recognition and Self-Supervised Models","code":"https://github.com/saurabh-kataria/PERL-samples","n_code_links":1,"syntology":null},{"rank_in_archive_order":25,"model":"PFPL","metrics":{"CBAK":"3.60","COVL":"3.67","CSIG":"4.18","PESQ (wb)":"3.15"},"uses_additional_data":false,"paper_date":"2020-10-28","paper":"/paper/improving-perceptual-quality-by-phone","paper_url":"https://arxiv.org/abs/2010.15174v3","paper_title":"Improving Perceptual Quality by Phone-Fortified Perceptual Loss using Wasserstein Distance for Speech Enhancement","code":"https://github.com/aleXiehta/PhoneFortifiedPerceptualLoss","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"MetricGAN+","metrics":{"CBAK":"3.16","COVL":"3.64","CSIG":"4.14","PESQ (wb)":"3.15"},"uses_additional_data":false,"paper_date":"2021-04-08","paper":"/paper/metricgan-an-improved-version-of-metricgan","paper_url":"https://arxiv.org/abs/2104.03538v2","paper_title":"MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement","code":"https://github.com/speechbrain/speechbrain/tree/develop/recipes/Voicebank/enhance/MetricGAN","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":11,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":27,"model":"MANNER-S + MV-AT (8.1GF)","metrics":{"CBAK":"3.61","COVL":"3.82","CSIG":"4.45","PESQ (wb)":"3.12","Para. (M)":"1.38","STOI":"95"},"uses_additional_data":false,"paper_date":"2022-08-22","paper":"/paper/multi-view-attention-transfer-for-efficient","paper_url":"https://arxiv.org/abs/2208.10367v2","paper_title":"Multi-View Attention Transfer for Efficient Speech Enhancement","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":28,"model":"MetricGAN-OKD (Causal Arch.)","metrics":{"CBAK":"3.13","COVL":"3.64","CSIG":"4.17","PESQ (wb)":"3.12","Para. (M)":"0.82"},"uses_additional_data":false,"paper_date":"2023-07-24","paper":"/paper/metricgan-okd-multi-metric-optimization-of","paper_url":"https://proceedings.mlr.press/v202/shin23b.html","paper_title":"MetricGAN-OKD: Multi-Metric Optimization of MetricGAN via Online Knowledge Distillation for Speech Enhancement","code":"https://github.com/wooseok-shin/MetricGAN-OKD","n_code_links":1,"syntology":null},{"rank_in_archive_order":29,"model":"SGMSE+","metrics":{"PESQ (wb)":"3.11"},"uses_additional_data":false,"paper_date":"2024-02-01","paper":"/paper/an-analysis-of-the-variance-of-diffusion","paper_url":"https://arxiv.org/abs/2402.00811v2","paper_title":"An Analysis of the Variance of Diffusion-based Speech Enhancement","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":30,"model":"DEMUCS (H=64, S=2 ,U =2)","metrics":{"CBAK":"3.4","COVL":"3.63","CSIG":"4.31","PESQ (wb)":"3.07","STOI":"95"},"uses_additional_data":false,"paper_date":"2020-06-23","paper":"/paper/real-time-speech-enhancement-in-the-waveform","paper_url":"https://arxiv.org/abs/2006.12847v3","paper_title":"Real Time Speech Enhancement in the Waveform Domain","code":"https://github.com/facebookresearch/denoiser","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":31,"model":"Dense-TSNet","metrics":{"CBAK":"3.58","COVL":"3.86","CSIG":"4.51","PESQ (wb)":"3.05","Para. (M)":"0.014"},"uses_additional_data":false,"paper_date":"2024-09-18","paper":"/paper/dense-tsnet-dense-connected-two-stage","paper_url":"https://arxiv.org/abs/2409.11725v1","paper_title":"Dense-TSNet: Dense Connected Two-Stage Structure for Ultra-Lightweight Speech Enhancement","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":32,"model":"RDL-Net 3.91M (Deep Xi - MMSE-LSA)","metrics":{"CBAK":"3.43","COVL":"3.72","CSIG":"4.38","PESQ (wb)":"3.02"},"uses_additional_data":false,"paper_date":"2020-02-27","paper":"/paper/deep-residual-dense-lattice-network-for","paper_url":"https://arxiv.org/abs/2002.12794v1","paper_title":"Deep Residual-Dense Lattice Network for Speech Enhancement","code":"https://github.com/nick-nikzad/RDL-SE","n_code_links":2,"syntology":null},{"rank_in_archive_order":33,"model":"ROSE","metrics":{"CBAK":"3.56","COVL":"3.72","CSIG":"4.47","PESQ (wb)":"3.01","Para. (M)":"36.98","STOI":"95"},"uses_additional_data":false,"paper_date":"2023-12-11","paper":"/paper/rose-a-recognition-oriented-speech","paper_url":"https://arxiv.org/abs/2312.06118v2","paper_title":"ROSE: A Recognition-Oriented Speech Enhancement Framework in Air Traffic Control Using Multi-Objective Learning","code":"https://github.com/xcyu-0903/rose","n_code_links":1,"syntology":null},{"rank_in_archive_order":34,"model":"FSPEN","metrics":{"PESQ (wb)":"2.97","Para. (M)":"0.079","STOI":"0.942"},"uses_additional_data":false,"paper_date":"2024-04-15","paper":"/paper/fspen-an-ultra-lightweight-network-for-real","paper_url":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10446016","paper_title":"FSPEN: AN ULTRA-LIGHTWEIGHT NETWORK FOR REAL TIME SPEECH ENAHNCMENT","code":"https://github.com/gitwukeyi/FSPEN","n_code_links":1,"syntology":null},{"rank_in_archive_order":35,"model":"RDL-Net 3.91M (Deep Xi - SRWF)","metrics":{"CBAK":"3.35","COVL":"3.67","CSIG":"4.36","PESQ (wb)":"2.94"},"uses_additional_data":false,"paper_date":"2020-02-27","paper":"/paper/deep-residual-dense-lattice-network-for","paper_url":"https://arxiv.org/abs/2002.12794v1","paper_title":"Deep Residual-Dense Lattice Network for Speech Enhancement","code":"https://github.com/nick-nikzad/RDL-SE","n_code_links":2,"syntology":null},{"rank_in_archive_order":36,"model":"RDL-Net 1.87M (Deep Xi - MMSE-LSA)","metrics":{"CBAK":"3.32","COVL":"3.62","CSIG":"4.29","PESQ (wb)":"2.93"},"uses_additional_data":false,"paper_date":"2020-02-27","paper":"/paper/deep-residual-dense-lattice-network-for","paper_url":"https://arxiv.org/abs/2002.12794v1","paper_title":"Deep Residual-Dense Lattice Network for Speech Enhancement","code":"https://github.com/nick-nikzad/RDL-SE","n_code_links":2,"syntology":null},{"rank_in_archive_order":37,"model":"Causal DEMUCS (H=48,S=4, U =4)","metrics":{"CBAK":"3.25","COVL":"3.52","CSIG":"4.22","PESQ (wb)":"2.93","STOI":"95"},"uses_additional_data":false,"paper_date":"2020-06-23","paper":"/paper/real-time-speech-enhancement-in-the-waveform","paper_url":"https://arxiv.org/abs/2006.12847v3","paper_title":"Real Time Speech Enhancement in the Waveform Domain","code":"https://github.com/facebookresearch/denoiser","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":38,"model":"SGMSE+ (Diffusion Model)","metrics":{"PESQ (wb)":"2.93"},"uses_additional_data":false,"paper_date":"2022-08-11","paper":"/paper/speech-enhancement-and-dereverberation-with","paper_url":"https://arxiv.org/abs/2208.05830v2","paper_title":"Speech Enhancement and Dereverberation with Diffusion-based Generative Models","code":"https://github.com/sp-uhh/sgmse","n_code_links":1,"syntology":null},{"rank_in_archive_order":39,"model":"MetricGAN","metrics":{"CBAK":"3.18","COVL":"3.42","CSIG":"3.99","PESQ (wb)":"2.86"},"uses_additional_data":false,"paper_date":"2019-05-13","paper":"/paper/metricgan-generative-adversarial-networks","paper_url":"https://arxiv.org/abs/1905.04874v1","paper_title":"MetricGAN: Generative Adversarial Networks based Black-box Metric Scores Optimization for Speech Enhancement","code":"https://github.com/anicolson/DeepXi","n_code_links":5,"syntology":null},{"rank_in_archive_order":40,"model":"RDL-Net 1.87M (Deep Xi - SRWF)","metrics":{"CBAK":"3.23","COVL":"3.56","CSIG":"4.27","PESQ (wb)":"2.84"},"uses_additional_data":false,"paper_date":"2020-02-27","paper":"/paper/deep-residual-dense-lattice-network-for","paper_url":"https://arxiv.org/abs/2002.12794v1","paper_title":"Deep Residual-Dense Lattice Network for Speech Enhancement","code":"https://github.com/nick-nikzad/RDL-SE","n_code_links":2,"syntology":null},{"rank_in_archive_order":41,"model":"real-time-GRU","metrics":{"PESQ (wb)":"2.82"},"uses_additional_data":false,"paper_date":"2021-02-15","paper":"/paper/a-modulation-domain-loss-for-neural-network-1","paper_url":"https://arxiv.org/abs/2102.07330v1","paper_title":"A Modulation-Domain Loss for Neural-Network-based Real-time Speech Enhancement","code":"https://github.com/tvuong123/ModulationDomainLoss","n_code_links":1,"syntology":null},{"rank_in_archive_order":42,"model":"DCT","metrics":{"CBAK":"3.29","COVL":"3.29","CSIG":"3.9","PESQ (wb)":"2.7"},"uses_additional_data":false,"paper_date":"2019-10-17","paper":"/paper/end-to-end-speech-enhancement-based-on","paper_url":"https://arxiv.org/abs/1910.07840v4","paper_title":"End-to-end speech enhancement based on discrete cosine transform","code":"https://github.com/BYRTIMO/END-TO-END-SPEECH-ENHANCEMENT-BASED-ON-DISCRETE-COSINE-TRANSFORM","n_code_links":3,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,795 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6795,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2785},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":7,"rows_with_any_sample_ran":5,"distinct_papers_with_graph_line":6,"distinct_papers_with_any_sample_ran":4,"samples_over_distinct_papers":{"n_ran":29,"n_unverified":20,"n_samples":49,"n_pointer_only_licence":16,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":30,"n_unverified":21,"n_samples":51,"n_pointer_only_licence":18,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}