mlsanityData to 19 Oct 2026

FA-GAN: Artifacts-free and Phase-aware High-fidelity GAN-based Vocoder

arXiv 2407.04575 ↗First posted 5 Jul 2024eess.AScited by 0

On the map: Speech and audio language models → Vocoders and speech synthesis

Abstract (arXiv)

Generative adversarial network (GAN) based vocoders have achieved significant attention in speech synthesis with high quality and fast inference speed. However, there still exist many noticeable spectral artifacts, resulting in the quality decline of synthesized speech. In this work, we adopt a novel GAN-based vocoder designed for few artifacts and high fidelity, called FA-GAN. To suppress the aliasing artifacts caused by non-ideal upsampling layers in high-frequency components, we introduce the anti-aliased twin deconvolution module in the generator. To alleviate blurring artifacts and enrich the reconstruction of spectral details, we propose a novel fine-grained multi-resolution real and imaginary loss to assist in the modeling of phase information. Experimental results reveal that FA-GAN outperforms the compared approaches in promoting audio quality and alleviating spectral artifacts, and exhibits superior performance when applied to unseen speaker scenarios.

Key facts. No citing papers in the map yet (data through 19 Oct 2026). Placed in Vocoders and speech synthesis, one of 11 directions in Speech and audio language models. Builds on 21 of its 30 references that are in the map. First posted on arXiv 5 Jul 2024.

Builds on

21 of 30 references.

Closest ancestors · 2

Same direction, newest first

  1. 2023iSTFTNet2: Faster and More Lightweight iSTFT-Based Neural Vocoder Using 1D-2D CNN14 Aug
  2. 2021SingGAN: Generative Adversarial Network For High-Fidelity Singing Voice Generation14 Oct

Other references · 19

  1. 2016WaveNet: A Generative Model for Raw Audiocited by 1,880
  2. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,462
  3. 2021Alias-Free Generative Adversarial Networkscited by 1,191
  4. 2020WaveGrad: Estimating Gradients for Waveform Generationcited by 645
  5. 2019MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesiscited by 436
  6. 2019Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogramcited by 425
  7. 2018WaveGlow: A Flow-based Generative Network for Speech Synthesiscited by 365
  8. 2022BigVGAN: A Universal Neural Vocoder with Large-Scale Trainingcited by 315
  9. 2017Parallel WaveNet: Fast High-Fidelity Speech Synthesiscited by 240
  10. 2016SampleRNN: An Unconditional End-to-End Neural Audio Generation Modelcited by 126
  11. 2020Neural Networks Fail to Learn Periodic Functions and How to Fix Itcited by 121
  12. 2021UnivNet: A Neural Vocoder with Multi-Resolution Spectrogram Discriminators for High-Fidelity Waveform Generationcited by 107
7 more
  1. 2020Multi-band MelGAN: Faster Waveform Generation for High-Quality Text-to-Speechcited by 91
  2. 2021PriorGrad: Improving Conditional Denoising Diffusion Models with Data-Dependent Adaptive Priorcited by 86
  3. 2022SpecGrad: Diffusion Probabilistic Model based Neural Vocoder with Adaptive Noise Spectral Shapingcited by 39
  4. 2020Upsampling artifacts in neural audio synthesiscited by 32
  5. 2021Fre-GAN: Adversarial Frequency-consistent Audio Synthesiscited by 30
  6. 2022Avocodo: Generative Adversarial Network for Artifact-free Vocodercited by 24
  7. 2020NanoFlow: Scalable Normalizing Flows with Sublinear Parameter Complexitycited by 14

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024VNet: A GAN-based Multi-Tier Discriminator Network for Speech Synthesis Vocoders
  2. 2024SpecDiff-GAN: A Spectrally-Shaped Noise Diffusion GAN for Speech and Music Synthesis
  3. 2023Source-Filter-Based Generative Adversarial Neural Vocoder for High Fidelity Speech Synthesis ↗
  4. 2026PhaseGAN: High-Fidelity Vocoder via Decoupled Amplitude and GAN-Driven Phase Reconstruction
  5. 2025DPN-GAN: Inducing Periodic Activations in Generative Adversarial Networks for High-Fidelity Audio Synthesis
  6. 2023APNet2: High-quality and High-efficiency Neural Vocoder with Direct Prediction of Amplitude and Phase Spectra
  7. 2022DSPGAN: a GAN-based universal vocoder for high-fidelity TTS by time-frequency domain supervision from DSP
  8. 2024JenGAN: Stacked Shifted Filters in GAN-Based Speech Synthesis