Methods › Audio › Generative Audio Models › WaveGAN
WaveGAN
Introduced by Chris Donahue et al. in Adversarial Audio Synthesis
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
WaveGAN is a generative adversarial network for unsupervised synthesis of raw-waveform audio (as opposed to image-like spectrograms).
The WaveGAN architecture is based off DCGAN. The DCGAN generator uses the transposed convolution operation to iteratively upsample low-resolution feature maps into a high-resolution image. WaveGAN modifies this transposed convolution operation to widen its receptive field, using a longer one-dimensional filters of length 25 instead of two-dimensional filters of size 5x5, and upsampling by a factor of 4 instead of 2 at each layer. The discriminator is modified in a similar way, using length-25 filters in one dimension and increasing stride from 2 to 4. These changes result in WaveGAN having the same number of parameters, numerical operations, and output dimensionality as DCGAN. An additional layer is added afterwards to allow for more audio samples. Further changes include:
- Flattening 2D convolutions into 1D (e.g. 5x5 2D conv becomes length-25 1D).
- Increasing the stride factor for all convolutions (e.g. stride 2x2 becomes stride 4).
- Removing batch normalization from the generator and discriminator.
- Training using the WGAN-GP strategy.
Papers archive 2025-07-28
30 shown of 30, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.
-
NAIST Simultaneous Speech Translation System for IWSLT 2024 30 Jun 2024 · 0 repositories · arXiv:2407.00826
-
(Un)paired signal-to-signal translation with 1D conditional GANs 5 Mar 2024 · 0 repositories · arXiv:2403.04800
-
The Effects of Signal-to-Noise Ratio on Generative Adversarial Networks Applied to Marine Bioacoustic Data 22 Dec 2023 · 0 repositories · arXiv:2312.14806
-
Framewise WaveGAN: High Speed Adversarial Vocoder in Time Domain with Very Low Computational Complexity 8 Dec 2022 · 0 repositories · arXiv:2212.04532
-
HiFi-WaveGAN: Generative Adversarial Network with Auxiliary Spectrogram-Phase Loss for High-Fidelity Singing Voice Generation 23 Oct 2022 · 1 repository · arXiv:2210.12740
-
WaveGAN: Frequency-aware GAN for High-Fidelity Few-shot Image Generation 15 Jul 2022 · 1 repository · arXiv:2207.07288Syntology ran 7 of 13 samples · 6 unverified · 13 pointer-only (licence)
-
WOLONet: Wave Outlooker for Efficient and High Fidelity Speech Synthesis 20 Jun 2022 · 0 repositories · arXiv:2206.09920
-
NatiQ: An End-to-end Text-to-Speech System for Arabic 15 Jun 2022 · 0 repositories · arXiv:2206.07373
-
Unified Source-Filter GAN with Harmonic-plus-Noise Source Excitation Generation 12 May 2022 · 0 repositories · arXiv:2205.06053
-
MSR-NV: Neural Vocoder Using Multiple Sampling Rates 28 Sep 2021 · 0 repositories · arXiv:2109.13714
-
Digital Einstein Experience: Fast Text-to-Speech for Conversational AI 21 Jul 2021 · 0 repositories · arXiv:2107.10658
-
StarGANv2-VC: A Diverse, Unsupervised, Non-parallel Framework for Natural-Sounding Voice Conversion 21 Jul 2021 · 2 repositories · arXiv:2107.10394Syntology ran 3 of 5 samples · 2 unverified
-
Interpreting intermediate convolutional layers of generative CNNs trained on waveforms 19 Apr 2021 · 0 repositories · arXiv:2104.09489
-
Unified Source-Filter GAN: Unified Source-filter Network Based On Factorization of Quasi-Periodic Parallel WaveGAN 10 Apr 2021 · 1 repository · arXiv:2104.04668
-
Adversarial Attacks and Defenses for Speech Recognition Systems 31 Mar 2021 · 0 repositories · arXiv:2103.17122
-
Improve GAN-based Neural Vocoder using Pointwise Relativistic LeastSquare GAN 26 Mar 2021 · 0 repositories · arXiv:2103.14245
-
LVCNet: Efficient Condition-Dependent Modeling Network for Waveform Generation 22 Feb 2021 · 5 repositories · arXiv:2102.10815
-
Study of Pre-processing Defenses against Adversarial Attacks on State-of-the-art Speaker Recognition Systems 22 Jan 2021 · 0 repositories · arXiv:2101.08909
-
Synthesising Realistic Calcium Imaging Data of Neuronal Populations Using GAN 1 Jan 2021 · 1 repository
-
StyleMelGAN: An Efficient High-Fidelity Adversarial Vocoder with Temporal Adaptive Normalization 3 Nov 2020 · 2 repositories · arXiv:2011.01557Syntology ran 0 of 2 samples · 2 unverified
-
Parallel waveform synthesis based on generative adversarial networks with voicing-aware conditional discriminators 27 Oct 2020 · 0 repositories · arXiv:2010.14151
-
Latent Vector Recovery of Audio GANs 16 Oct 2020 · 0 repositories · arXiv:2010.08534
-
Baseline System of Voice Conversion Challenge 2020 with Cyclic Variational Autoencoder and Parallel WaveGAN 9 Oct 2020 · 1 repository · arXiv:2010.04429
-
Synthesising Realistic Calcium Traces of Neuronal Populations Using GAN 6 Sep 2020 · 1 repository · arXiv:2009.02707
-
HiFiSinger: Towards High-Fidelity Neural Singing Voice Synthesis 3 Sep 2020 · 1 repository · arXiv:2009.01776Syntology ran 0 of 10 samples · 10 unverified
-
VocGAN: A High-Fidelity Real-time Vocoder with a Hierarchically-nested Adversarial Network 30 Jul 2020 · 2 repositories · arXiv:2007.15256
-
Quasi-Periodic Parallel WaveGAN: A Non-autoregressive Raw Waveform Generative Model with Pitch-dependent Dilated Convolution Neural Network 25 Jul 2020 · 1 repository · arXiv:2007.12955
-
Towards Robust Neural Vocoding for Speech Generation: A Survey 5 Dec 2019 · 0 repositories · arXiv:1912.02461
-
Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogram 25 Oct 2019 · 12 repositories · arXiv:1910.11480Syntology ran 0 of 20 samples · 20 unverified · 1 pointer-only (licence)
-
Adversarial Audio Synthesis 12 Feb 2018 · 22 repositories · arXiv:1802.04208Syntology ran 2 of 4 samples · 2 unverified · 4 pointer-only (licence)
Tasks archive 2025-07-28
20 shown of 29 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
| Task | Papers |
|---|---|
| Text to Speech | 9 |
| text-to-speech | 9 |
| Generative Adversarial Network | 7 |
| Speech Synthesis | 4 |
| Vocal Bursts Intensity Prediction | 4 |
| Image Generation | 3 |
| Voice Conversion | 3 |
| Audio Generation | 2 |
| CPU | 2 |
| Decoder | 2 |
| GPU | 2 |
| Singing Voice Synthesis | 2 |
| Translation | 2 |
| Adversarial Robustness | 1 |
| Audio Synthesis | 1 |
| Automatic Speech Recognition | 1 |
| Automatic Speech Recognition (ASR) | 1 |
| Image-to-Image Translation | 1 |
| Speaker Recognition | 1 |
| Spectral Reconstruction | 1 |
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections