mlsanityData to 14 Oct 2026

FlashSR: One-step Versatile Audio Super-resolution via Diffusion Distillation

arXiv 2501.10807 ↗First posted 18 Jan 2025eess.AS cs.SDcited by 5

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

Abstract (arXiv)

Versatile audio super-resolution (SR) is the challenging task of restoring high-frequency components from low-resolution audio with sampling rates between 4kHz and 32kHz in various domains such as music, speech, and sound effects. Previous diffusion-based SR methods suffer from slow inference due to the need for a large number of sampling steps. In this paper, we introduce FlashSR, a single-step diffusion model for versatile audio super-resolution aimed at producing 48kHz audio. FlashSR achieves fast inference by utilizing diffusion distillation with three objectives: distillation loss, adversarial loss, and distribution-matching distillation loss. We further enhance performance by proposing the SR Vocoder, which is specifically designed for SR models operating on mel-spectrograms. FlashSR demonstrates competitive performance with the current state-of-the-art model in both objective and subjective evaluations while being approximately 22 times faster.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Placed in Audio, speech and music, one of 14 directions in Diffusion models. Builds on 32 of its 43 references that are in the map. First posted on arXiv 18 Jan 2025.

Builds on

32 of 43 references.

Closest ancestors · 6

Same direction, newest first

  1. 2024AudioLCM: Text-to-Audio Generation with Latent Consistency Models1 Jun
  2. 2024Music Consistency Models20 Apr
  3. 2023ConsistencyTTA: Accelerating Diffusion-Based Text-to-Audio Generation with Consistency Distillation19 Sep
  4. 2023CoMoSpeech: One-Step Speech and Singing Voice Synthesis via Consistency Model11 May
  5. 2022ProDiff: Progressive Fast Diffusion Model For High-Quality Text-to-Speech13 Jul
  6. 2021NU-Wave: A Diffusion Probabilistic Model for Neural Audio Upsampling6 Apr

Other references · 26

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  4. 2020Denoising Diffusion Implicit Modelscited by 9,781
  5. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  6. 2022Classifier-Free Diffusion Guidancecited by 5,847
  7. 2016Gaussian Error Linear Units (GELUs)cited by 3,502
  8. 2018Group Normalizationcited by 1,524
  9. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,154
  10. 2023One-step Diffusion with Distribution Matching Distillationcited by 1,019
  11. 2022DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Modelscited by 938
  12. 2021Tackling the Generative Learning Trilemma with Denoising Diffusion GANscited by 516
14 more
  1. 2023High-Fidelity Audio Compression with Improved RVQGANcited by 431
  2. 2015Autoencoding beyond pixels using a learned similarity metriccited by 407
  3. 2023WavCaps: A ChatGPT-Assisted Weakly-Labelled Audio Captioning Dataset for Audio-Language Multimodal Researchcited by 267
  4. 2022BigVGAN: A Universal Neural Vocoder with Large-Scale Trainingcited by 251
  5. 2016FMA: A Dataset For Music Analysiscited by 207
  6. 2023UFOGen: You Forward Once Large Scale Text-to-Image Generation via Diffusion GANscited by 166
  7. 2021UnivNet: A Neural Vocoder with Multi-Resolution Spectrogram Discriminators for High-Fidelity Waveform Generationcited by 74
  8. 2024Flash Diffusion: Accelerating Any Conditional Diffusion Model for Few Steps Image Generationcited by 56
  9. 2021VoiceFixer: Toward General Speech Restoration with Neural Vocodercited by 50
  10. 2023Moisesdb: A dataset for source separation beyond 4-stemscited by 39
  11. 2023AudioSR: Versatile Audio Super-resolution at Scalecited by 39
  12. 2022Neural Vocoder is All You Need for Speech Super-resolutioncited by 37
  13. 2023Multi-Scale Sub-Band Constant-Q Transform Discriminator for High-Fidelity Vocodercited by 13
  14. 2020Phase-aware music super-resolution using generative adversarial networkscited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 3.

5 papers.

  1. 2025Audio Super-Resolution with Latent Bridge Modelscited by 14
  2. 2025UniverSR: Unified and Versatile Audio Super-Resolution via Vocoder-Free Flow Matchingcited by 3
  3. 2026LatentFlowSR: High-Fidelity Audio Super-Resolution via Noise-Robust Latent Flow Matchingcited by 1
  4. 2026Discriminating real and synthetic super-resolved audio samples using embedding-based classifierscited by 0
  5. 2026VeRe-Flow: Guiding Flow Matching toward Clean Speech via Velocity Contrastive Regularization and Representation Alignment for Noise-Robust Bandwidth Expansioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026FastWave: Optimized Diffusion Model for Audio Super-Resolution
  2. 2025Inference-time Scaling for Diffusion-based Audio Super-resolution
  3. 2025SAGA-SR: Semantically and Acoustically Guided Audio Super-Resolution
  4. 2025FLowHigh: Towards Efficient and High-Quality Audio Super-Resolution with Single-Step Flow Matching
  5. 2022Conditioning and Sampling in Variational Diffusion Models for Speech Super-Resolution
  6. 2026A Survey of Advancing Audio Super-Resolution and Bandwidth Extension from Discriminative to Generative Models
  7. 2022NU-Wave 2: A General Neural Audio Upsampling Model for Various Sampling Rates
  8. 2025Score Distillation Sampling for Audio: Source Separation, Synthesis, and Beyond