mlsanityData to 7 Oct 2026

GAN You Hear Me? Reclaiming Unconditional Speech Synthesis from Diffusion Models

arXiv 2210.05271 ↗First posted 11 Oct 2022cs.SD cs.AI eess.AScited by 3

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

Abstract (arXiv)

We propose AudioStyleGAN (ASGAN), a new generative adversarial network (GAN) for unconditional speech synthesis. As in the StyleGAN family of image synthesis models, ASGAN maps sampled noise to a disentangled latent vector which is then mapped to a sequence of audio features so that signal aliasing is suppressed at every layer. To successfully train ASGAN, we introduce a number of new techniques, including a modification to adaptive discriminator augmentation to probabilistically skip discriminator updates. ASGAN achieves state-of-the-art results in unconditional speech synthesis on the Google Speech Commands dataset. It is also substantially faster than the top-performing diffusion models. Through a design that encourages disentanglement, ASGAN is able to perform voice conversion and speech editing without being explicitly trained to do so. ASGAN demonstrates that GANs are still highly competitive with diffusion models. Code, models, samples: https://github.com/RF5/simple-asgan/.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Placed in Audio, speech and music, one of 14 directions in Diffusion models. Builds on 31 of its 35 references that are in the map. First posted on arXiv 11 Oct 2022.

Builds on

31 of 35 references.

Closest ancestors · 1

Same direction, newest first

  1. 2020DiffWave: A Versatile Diffusion Model for Audio Synthesis21 Sep

Other references · 30

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  3. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  4. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  5. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  6. 2022Generative Adversarial Networkscited by 5,603
  7. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  8. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  9. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  10. 2016Improved Techniques for Training GANscited by 2,938
  11. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
  12. 2021HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Unitscited by 2,301
18 more
  1. 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
  2. 2016WaveNet: A Generative Model for Raw Audiocited by 1,710
  3. 2020Training Generative Adversarial Networks with Limited Datacited by 1,184
  4. 2021Alias-Free Generative Adversarial Networkscited by 1,181
  5. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,139
  6. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 741
  7. 2018Which Training Methods for GANs do actually Converge?cited by 672
  8. 2021Generative Spoken Language Modeling from Raw Audiocited by 336
  9. 2021Speech Resynthesis from Discrete Disentangled Self-Supervised Representationscited by 243
  10. 2018Adversarial Audio Synthesiscited by 222
  11. 2022It's Raw! Audio Generation with State-Space Modelscited by 171
  12. 2021Text-Free Prosody-Aware Generative Spoken Language Modelingcited by 119
  13. 2017DeLiGAN : Generative Adversarial Networks for Diverse and Limited Datacited by 77
  14. 2022The VoiceMOS Challenge 2022cited by 57
  15. 2020Generative Adversarial Phonology: Modeling unsupervised phonetic and phonological learning with neural networkscited by 12
  16. 2019Alleviation of Gradient Exploding in GANs: Fake Can Be Realcited by 8
  17. 2020Local and non-local dependency learning and emergence of rule-like representations in speech data by Deep Convolutional Generative Adversarial Networkscited by 7
  18. 2022Generating Diverse Vocal Bursts with StyleGAN2 and MEL-Spectrogramscited by 3

Built on by

Cited by year

Citing papers per year: 2023: 3.

3 papers.

  1. 2023StyleTTS 2: Towards Human-Level Text-to-Speech through Style Diffusion and Adversarial Training with Large Speech Language Modelscited by 148
  2. 2023EDMSound: Spectrogram Based Diffusion Models for Efficient and High-Quality Audio Synthesiscited by 15
  3. 2023Disentanglement in a GAN for Unconditional Speech Synthesiscited by 4

Similar, not cited

Close in wording, no citation either way

  1. 2023Voice Conversion with Denoising Diffusion Probabilistic GAN Models
  2. 2022DiffGAN-TTS: High-Fidelity and Efficient Text-to-Speech with Denoising Diffusion GANs
  3. 2025Flow2GAN: Hybrid Flow Matching and GAN with Multi-Resolution Network for Few-step High-Fidelity Audio Generation
  4. 2023Adversarial Training of Denoising Diffusion Model Using Dual Discriminators for High-Fidelity Multi-Speaker TTS
  5. 2025Leveraging Discriminative Latent Representations for Conditioning GAN-Based Speech Enhancement
  6. 2025On the Design of Diffusion-based Neural Speech Codecs
  7. 2022Mandarin Singing Voice Synthesis with Denoising Diffusion Probabilistic Wasserstein GAN
  8. 2026WaveNeXt 2: ConvNeXt-Based Fast Neural Vocoders With Residual Denoising and Sub-Modeling for GAN and Diffusion Models