mlsanityData to 7 Oct 2026

CMGAN: Conformer-based Metric GAN for Speech Enhancement

arXiv 2203.15149 ↗First posted 28 Mar 2022cs.SD cs.AI cs.LGcited by 45

Not on the map. Why?

Abstract (arXiv)

Recently, convolution-augmented transformer (Conformer) has achieved promising performance in automatic speech recognition (ASR) and time-domain speech enhancement (SE), as it can capture both local and global dependencies in the speech signal. In this paper, we propose a conformer-based metric generative adversarial network (CMGAN) for SE in the time-frequency (TF) domain. In the generator, we utilize two-stage conformer blocks to aggregate all magnitude and complex spectrogram information by modeling both time and frequency dependencies. The estimation of magnitude and complex spectrogram is decoupled in the decoder stage and then jointly incorporated to reconstruct the enhanced speech. In addition, a metric discriminator is employed to further improve the quality of the enhanced estimated speech by optimizing the generator with respect to a corresponding evaluation score. Quantitative analysis on Voice Bank+DEMAND dataset indicates the capability of CMGAN in outperforming various previous models with a margin, i.e., PESQ of 3.41 and SSNR of 11.10 dB.

Key facts. Cited by 45 papers in the map (data through 7 Oct 2026). Builds on 24 of its 40 references that are in the map. First posted on arXiv 28 Mar 2022.

Builds on

24 of 40 references.

References · 24

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  3. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  4. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,838
  5. 2016Instance Normalization: The Missing Ingredient for Fast Stylizationcited by 1,187
  6. 2016Least Squares Generative Adversarial Networkscited by 1,000
  7. 2017Supervised Speech Separation Based on Deep Learning: An Overviewcited by 213
  8. 2020Real Time Speech Enhancement in the Waveform Domaincited by 183
  9. 2017SEGAN: Speech Enhancement Generative Adversarial Networkcited by 165
  10. 2021MetricGAN+: An Improved Version of MetricGAN for Speech Enhancementcited by 114
  11. 2019MetricGAN: Generative Adversarial Networks based Black-box Metric Scores Optimization for Speech Enhancementcited by 104
  12. 2020Dual-Path Transformer Network: Direct Context-Aware Modeling for End-to-End Monaural Speech Separationcited by 67
12 more
  1. 2022ICASSP 2022 Deep Noise Suppression Challengecited by 57
  2. 2019PHASEN: A Phase-and-Harmonics-Aware Speech Enhancement Networkcited by 49
  3. 2020Continuous Speech Separation with Conformercited by 36
  4. 2021TSTNN: Two-stage Transformer based Neural Network for Speech Enhancement in the Time Domaincited by 36
  5. 2021DPT-FSNet: Dual-path Transformer Based Full-band and Sub-band Fusion Network for Speech Enhancementcited by 33
  6. 2021Glance and Gaze: A Collaborative Learning Framework for Single-channel Speech Enhancementcited by 25
  7. 2018Improved Speech Enhancement with the Wave-U-Netcited by 23
  8. 2020A consolidated view of loss functions for supervised deep learning-based speech enhancementcited by 19
  9. 2021Dual-branch Attention-In-Attention Transformer for single-channel speech enhancementcited by 18
  10. 2021On The Compensation Between Magnitude and Phase in Speech Separationcited by 12
  11. 2018Exploring Tradeoffs in Models for Low-latency Speech Enhancementcited by 8
  12. 2020Investigating Cross-Domain Losses for Speech Enhancementcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 7; 2023: 7; 2024: 11; 2025: 12; 2026: 8.

45 papers.

Sort:

By area:

  1. 2022Speech Enhancement and Dereverberation with Diffusion-based Generative Modelscited by 195
  2. 2023LauraGPT: Listen, Attend, Understand, and Regenerate Audio with GPTcited by 77
  3. 2023Separate Anything You Describecited by 43
  4. 2022CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancementcited by 22
  5. 2024The PESQetarian: On the Relevance of Goodhart's Law for Speech Enhancementcited by 12
  6. 2024Mamba-SEUNet: Mamba UNet for Monaural Speech Enhancementcited by 10
  7. 2022SCP-GAN: Self-Correcting Discriminator Optimization for Training Consistency Preserving Metric GAN on Speech Enhancement Taskscited by 8
  8. 2023Noise robust speech emotion recognition with signal-to-noise ratio adapting speech enhancementcited by 5
  9. 2024A Variance-Preserving Interpolation Approach for Diffusion Models with Applications to Single Channel Speech Enhancement and Recognitioncited by 5
  10. 2024MUSE: Flexible Voiceprint Receptive Fields and Multi-Path Fusion Enhanced Taylor Transformer for U-Net-based Speech Enhancementcited by 4
  11. 2022Deep neural network techniques for monaural speech enhancement: state of the art analysiscited by 3
  12. 2023Large-scale unsupervised audio pre-training for video-to-speech synthesiscited by 3
33 more
  1. 2025A Survey of Deep Learning for Complex Speech Spectrogramscited by 3
  2. 2023Multi-Dimensional and Multi-Scale Modeling for Speech Separation Optimized by Discriminative Learningcited by 2
  3. 2023DPATD: Dual-Phase Audio Transformer for Denoisingcited by 2
  4. 2024From KAN to GR-KAN: Advancing Speech Enhancement with KAN-Based Methodologycited by 2
  5. 2025A Two-Stage Hierarchical Deep Filtering Framework for Real-Time Speech Enhancementcited by 2
  6. 2022A Training and Inference Strategy Using Noisy and Enhanced Speech as Target for Speech Enhancement without Clean Speechcited by 1
  7. 2022Audio Denoising for Robust Audio Fingerprintingcited by 1
  8. 2024Diffusion Gaussian Mixture Audio Denoisecited by 1
  9. 2024Complex Image-Generative Diffusion Transformer for Audio Denoisingcited by 1
  10. 2024Effective Noise-aware Data Simulation for Domain-adaptive Speech Enhancement Leveraging Dynamic Stochastic Perturbationcited by 1
  11. 2025Collective Learning Mechanism based Optimal Transport Generative Adversarial Network for Non-parallel Voice Conversioncited by 1
  12. 2025DeepFilterGAN: A Full-band Real-time Speech Enhancement System with GAN-based Stochastic Regenerationcited by 1
  13. 2025Improved Intelligibility of Dysarthric Speech using Conditional Flow Matchingcited by 1
  14. 2025WhiSQA: Non-Intrusive Speech Quality Prediction Using Whisper Encoder Featurescited by 1
  15. 2025Leveraging Discriminative Latent Representations for Conditioning GAN-Based Speech Enhancementcited by 1
  16. 2026Quaternion Self-Attention with Shared Scorescited by 1
  17. 2022Inference and Denoise: Causal Inference-based Neural Speech Enhancementcited by 0
  18. 2023Music Augmentation and Denoising For Peak-Based Audio Fingerprintingcited by 0
  19. 2024Objective and subjective evaluation of speech enhancement methods in the UDASE task of the 7th CHiME challengecited by 0
  20. 2024TRNet: Two-level Refinement Network leveraging Speech Enhancement for Noise Robust Speech Emotion Recognitioncited by 0
  21. 2024A Dual-Branch Parallel Network for Speech Enhancement and Restorationcited by 0
  22. 2025Single-Channel Distance-Based Source Separation for Mobile GPU in Outdoor and Indoor Environmentscited by 0
  23. 2025Active Speech Enhancement: Active Speech Denoising Decliping and Deveraberationcited by 0
  24. 2025EffiFusion-GAN: Efficient Fusion Generative Adversarial Network for Speech Enhancementcited by 0
  25. 2025ADNAC: Audio Denoiser using Neural Audio Codeccited by 0
  26. 2025Lightweight Wasserstein Audio-Visual Model for Unified Speech Enhancement and Separationcited by 0
  27. 2026Universal Robust Speech Adaptation for Cross-Domain Speech Recognition and Enhancementcited by 0
  28. 2026LLM-Guided Reinforcement Learning for Audio-Visual Speech Enhancementcited by 0
  29. 2026BASENet: Band-Adapted Speech Enhancement Network with Cross-Band Attentioncited by 0
  30. 2026QC-GAN: A Parameter-Efficient Quaternion Conformer GAN for High-Fidelity Speech Enhancementcited by 0
  31. 2026StrixAE: An Intelligent Agent for Audio Enhancement under Complex Distortion Coupling in Real-World Scenarioscited by 0
  32. 2026Beyond Model Size: Redesigning LiSenNet for embedded speech enhancementcited by 0
  33. 2026Perceptual Quality Loss or Loss of Perceptual Quality?cited by 0