mlsanityData to 14 Oct 2026

Music Separation Enhancement with Generative Modeling

arXiv 2208.12387 ↗First posted 26 Aug 2022cs.SD cs.LG eess.AScited by 8

Not on the map. Why?

Abstract (arXiv)

Despite phenomenal progress in recent years, state-of-the-art music separation systems produce source estimates with significant perceptual shortcomings, such as adding extraneous noise or removing harmonics. We propose a post-processing model (the Make it Sound Good (MSG) post-processor) to enhance the output of music source separation systems. We apply our post-processing model to state-of-the-art waveform-based and spectrogram-based music source separators, including a separator unseen by MSG during training. Our analysis of the errors produced by source separators shows that waveform models tend to introduce more high-frequency noise, while spectrogram models tend to lose transients and high frequency content. We introduce objective measures to quantify both kinds of errors and show MSG improves the source reconstruction of both kinds of errors. Crowdsourced subjective evaluations demonstrate that human listeners prefer source estimates of bass and drums that have been post-processed by MSG.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 39 of its 63 references that are in the map. First posted on arXiv 26 Aug 2022.

Builds on

39 of 63 references.

References · 39

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,154
  3. 2016Least Squares Generative Adversarial Networkscited by 1,001
  4. 2018SDR - half-baked or well done?cited by 423
  5. 2015Autoencoding beyond pixels using a learned similarity metriccited by 407
  6. 2019MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesiscited by 354
  7. 2019Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogramcited by 340
  8. 2018Adversarial Audio Synthesiscited by 225
  9. 2015Deep clustering: Discriminative embeddings for segmentation and separationcited by 212
  10. 2020Real Time Speech Enhancement in the Waveform Domaincited by 186
  11. 2017SEGAN: Speech Enhancement Generative Adversarial Networkcited by 166
  12. 2018Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separationcited by 125
27 more
  1. 2019GANSynth: Adversarial Neural Audio Synthesiscited by 116
  2. 2017TasNet: time-domain audio separation network for real-time, single-channel speech separationcited by 115
  3. 2021MetricGAN+: An Improved Version of MetricGAN for Speech Enhancementcited by 114
  4. 2019MetricGAN: Generative Adversarial Networks based Black-box Metric Scores Optimization for Speech Enhancementcited by 104
  5. 2019Music Source Separation in the Waveform Domaincited by 99
  6. 2019High Fidelity Speech Synthesis with Adversarial Networkscited by 92
  7. 2021Hybrid Spectrogram and Waveform Source Separationcited by 91
  8. 2021UnivNet: A Neural Vocoder with Multi-Resolution Spectrogram Discriminators for High-Fidelity Waveform Generationcited by 74
  9. 2019Neural source-filter waveform models for statistical parametric speech synthesiscited by 46
  10. 2018The 2018 Signal Separation Evaluation Campaigncited by 41
  11. 2020HiFi-GAN: High-Fidelity Denoising and Dereverberation Based on Speech Deep Features in Adversarial Networkscited by 40
  12. 2021Music Demixing Challenge 2021cited by 38
  13. 2020Upsampling artifacts in neural audio synthesiscited by 28
  14. 2020D3Net: Densely connected multidilated DenseNet for music source separationcited by 24
  15. 2018End-to-end music source separation: is it possible in the waveform domain?cited by 19
  16. 2019Simultaneous Separation and Transcription of Mixtures with Multiple Polyphonic and Percussive Instrumentscited by 18
  17. 2022On loss functions and evaluation metrics for music source separationcited by 17
  18. 2022A Two-Stage U-Net for High-Fidelity Denoising of Historical Recordingscited by 17
  19. 2017Adversarial Semi-Supervised Audio Source Separation applied to Singing Voice Extractioncited by 11
  20. 2022Music Enhancement via Image Translation and Vocodingcited by 11
  21. 2016Deep Clustering and Conventional Networks for Music Separation: Stronger Togethercited by 10
  22. 2019Bandwidth Extension on Raw Audio via Generative Adversarial Networkscited by 10
  23. 2020On Filter Generalization for Music Bandwidth Extension Using Deep Neural Networkscited by 10
  24. 2021CWS-PResUNet: Music Source Separation with Channel-wise Subband Phase-aware ResUNetcited by 9
  25. 2020Learning to Denoise Historical Musiccited by 8
  26. 2021Upsampling layers for music source separationcited by 2
  27. 2022Reproducible Subjective Evaluationcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 3; 2024: 1; 2025: 1; 2026: 2.

8 papers.

  1. 2023TokenSplit: Using Discrete Speech Representations for Direct, Refined, and Transcript-Conditioned Speech Separation and Recognitioncited by 8
  2. 2022JaCappella Corpus: A Japanese a Cappella Vocal Ensemble Corpuscited by 5
  3. 2023Exploiting Time-Frequency Conformers for Music Audio Enhancementcited by 2
  4. 2023Distribution Preserving Source Separation With Time Frequency Predictive Modelscited by 1
  5. 2024Improving Music Source Separation with Diffusion and Consistency Refinementcited by 1
  6. 2025Generating Separated Singing Vocals Using a Diffusion Model Conditioned on Music Mixturescited by 1
  7. 2026A Knowledge-Driven Approach to Music Segmentation, Music Source Separation and Cinematic Audio Source Separationcited by 0
  8. 2026DTT-BSR+: A Generative-Regression Cascade for Music Source Restorationcited by 0