mlsanityData to 7 Oct 2026

Absorbing Discrete Diffusion for Speech Enhancement

arXiv 2602.22417 ↗First posted 25 Feb 2026cs.SD eess.AScited by 0

On the map: Diffusion models → Audio, speech and music

Abstract (arXiv)

Inspired by recent developments in neural speech coding and diffusion-based language modeling, we tackle speech enhancement by modeling the conditional distribution of clean speech codes given noisy speech codes using absorbing discrete diffusion. The proposed approach, which we call ADDSE, leverages both the expressive latent space of neural audio codecs and the non-autoregressive sampling procedure of diffusion models. To efficiently model the hierarchical structure of residual vector quantization codes, we propose RQDiT, which combines techniques from RQ-Transformer and diffusion Transformers for non-autoregressive modeling. Results show competitive performance in terms of non-intrusive objective metrics on two datasets, especially at low signal-to-noise ratios and with few sampling steps. Code and audio examples are available online.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Audio, speech and music, one of 14 directions in Diffusion models. Builds on 63 of its 79 references that are in the map. First posted on arXiv 25 Feb 2026.

Builds on

63 of 79 references.

Closest ancestors · 9

Same direction, newest first

  1. 2025DiTSE: High-Fidelity Generative Speech Enhancement via Latent Diffusion Transformers13 Apr
  2. 2025ProSE: Diffusion Priors for Speech Enhancement9 Mar
  3. 2024High-Resolution Speech Restoration with Latent Diffusion Model17 Sep
  4. 2024The Effect of Training Dataset Size on Discriminative and Diffusion-Based Speech Enhancement Systems10 Jun
  5. 2024Diffusion Models for Audio Restoration15 Feb
  6. 2023Investigating the Design Space of Diffusion Models for Speech Enhancement7 Dec
  7. 2023Diffusion-Based Speech Enhancement in Matched and Mismatched Conditions Using a Heun-Based Sampler5 Dec
  8. 2022Speech Enhancement and Dereverberation with Diffusion-based Generative Models11 Aug
  9. 2022Conditional Diffusion Probabilistic Model for Speech Enhancement10 Feb

Other references · 54

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  4. 2022Scalable Diffusion Models with Transformerscited by 5,931
  5. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  6. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 2,994
  7. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,286
  8. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,665
  9. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,139
  10. 2022MaskGIT: Masked Generative Image Transformercited by 1,082
  11. 2023PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesiscited by 854
  12. 2021SoundStream: An End-to-End Neural Audio Codeccited by 756
38 more
  1. 2023Discrete Diffusion Modeling by Estimating the Ratios of the Data Distributioncited by 712
  2. 2022Autoregressive Image Generation using Residual Quantizationcited by 706
  3. 2022High Fidelity Neural Audio Compressioncited by 673
  4. 2021Vector-quantized Image Modeling with Improved VQGANcited by 656
  5. 2024Simplified and Generalized Masked Diffusion for Discrete Datacited by 606
  6. 2024Moshi: a speech-text foundation model for real-time dialoguecited by 536
  7. 2020MLS: A Large-Scale Multilingual Dataset for Speech Researchcited by 467
  8. 2022A Continuous Time Framework for Discrete Denoising Modelscited by 437
  9. 2023High-Fidelity Audio Compression with Improved RVQGANcited by 423
  10. 2018Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separationcited by 400
  11. 2023Analyzing and Improving the Training Dynamics of Diffusion Modelscited by 396
  12. 2024Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Datacited by 371
  13. 2020FSD50K: An Open Dataset of Human-Labeled Sound Eventscited by 339
  14. 2016FMA: A Dataset For Music Analysiscited by 205
  15. 2021NISQA: A Deep CNN-Self-Attention Model for Multidimensional Speech Quality Prediction with Crowdsourced Datasetscited by 189
  16. 2022Score-based Continuous-time Discrete Diffusion Modelscited by 167
  17. 2021DNSMOS P.835: A Non-Intrusive Perceptual Objective Speech Quality Metric to Evaluate Noise Suppressorscited by 165
  18. 2019WHAM!: Extending Speech Separation to Noisy Environmentscited by 154
  19. 2022Concrete Score Matching: Generalized Score Matching for Discrete Datacited by 144
  20. 2020Neural Networks Fail to Learn Periodic Functions and How to Fix Itcited by 97
  21. 2024The T05 System for The VoiceMOS Challenge 2024: Transfer Learning from Deep Image Classifier to Naturalness MOS Prediction of High-Quality Synthetic Speechcited by 78
  22. 2020Wavesplit: End-to-End Speech Separation by Speaker Clusteringcited by 74
  23. 2022Music Source Separation with Band-split RNNcited by 69
  24. 2023ICASSP 2023 Deep Noise Suppression Challengecited by 62
  25. 2024EARS: An Anechoic Fullband Speech Dataset Benchmarked for Speech Enhancement and Dereverberationcited by 62
  26. 2024BigCodec: Pushing the Limits of Low-Bitrate Neural Speech Codeccited by 45
  27. 2024SpeechBERTScore: Reference-Aware Automatic Evaluation of Speech Generation Leveraging NLP Evaluation Metricscited by 38
  28. 2023SELM: Speech Enhancement Using Discrete Tokens and Language Modelscited by 36
  29. 2024MaskSR: Masked Language Model for Full-band Speech Restorationcited by 24
  30. 2024SCOREQ: Speech Quality Assessment with Contrastive Regressioncited by 23
  31. 2025GenSE: Generative Speech Enhancement via Language Models using Hierarchical Modelingcited by 20
  32. 2025LLaSE-G1: Incentivizing Generalization Capability for LLaMA-based Speech Enhancementcited by 19
  33. 2025FocalCodec: Low-Bitrate Speech Coding via Focal Modulation Networkscited by 18
  34. 2023Low-latency Speech Enhancement via Speech Token Generationcited by 7
  35. 2025Speech Enhancement Using Continuous Embeddings of Neural Audio Codeccited by 7
  36. 2025Modeling strategies for speech enhancement in the latent space of a neural audio codeccited by 6
  37. 2024Joint Semantic Knowledge Distillation and Masked Acoustic Modeling for Full-band Speech Restoration with Improved Intelligibilitycited by 5
  38. 2025UniSE: A Unified Framework for Decoder-Only Autoregressive LM-Based Speech Enhancementcited by 4

50 most cited of 54 All 54 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Posterior Transition Modeling for Unsupervised Diffusion-Based Speech Enhancement
  2. 2023Single and Few-step Diffusion for Generative Speech Enhancement
  3. 2026Rethinking Language Model-Based Generative Speech Enhancement in the Latent Space of a Neural Audio Codec
  4. 2023Generative De-Quantization for Neural Speech Codec via Latent Diffusion
  5. 2023Diffusion-based speech enhancement with a weighted generative-supervised learning loss
  6. 2023A Survey on Audio Diffusion Models: Text To Speech Synthesis and Enhancement in Generative AI
  7. 2023Unsupervised speech enhancement with diffusion-based generative models
  8. 2024Pre-training Feature Guided Diffusion Model for Speech Enhancement