Vocoder-Projected Feature Discriminator
Not on the map. Why?
Abstract (arXiv)
In text-to-speech (TTS) and voice conversion (VC), acoustic features, such as mel spectrograms, are typically used as synthesis or conversion targets owing to their compactness and ease of learning. However, because the ultimate goal is to generate high-quality waveforms, employing a vocoder to convert these features into waveforms and applying adversarial training in the time domain is reasonable. Nevertheless, upsampling the waveform introduces significant time and memory overheads. To address this issue, we propose a vocoder-projected feature discriminator (VPFD), which uses vocoder features for adversarial training. Experiments on diffusion-based VC distillation demonstrated that a pretrained and frozen vocoder feature extractor with a single upsampling step is necessary and sufficient to achieve a VC performance comparable to that of waveform discriminators while reducing the training time and memory consumption by 9.6 and 11.4 times, respectively.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 38 of its 45 references that are in the map. First posted on arXiv 25 Aug 2025.
Builds on
38 of 45 references.
References · 38
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2014Adam: A Method for Stochastic Optimizationcited by 37,240
- 2020Denoising Diffusion Probabilistic Modelscited by 21,076
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
- 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
- 2022Robust Speech Recognition via Large-Scale Weak Supervisioncited by 4,057
- 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,154
- 2016Least Squares Generative Adversarial Networkscited by 1,001
- 2023Adversarial Diffusion Distillationcited by 759
- 2016Language Modeling with Gated Convolutional Networkscited by 702
- 2019LibriTTS: A Corpus Derived from LibriSpeech for Text-to-Speechcited by 649
- 2016Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networkscited by 543
26 more
- 2022StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasetscited by 492
- 2015Autoencoding beyond pixels using a learned similarity metriccited by 407
- 2022UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022cited by 390
- 2019MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesiscited by 354
- 2019Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogramcited by 340
- 2018Transfer Learning from Speaker Verification to Multispeaker Text-To-Speech Synthesiscited by 282
- 2022BigVGAN: A Universal Neural Vocoder with Large-Scale Trainingcited by 251
- 2020DNSMOS: A Non-Intrusive Perceptual Objective Speech Quality metric to evaluate Noise Suppressorscited by 241
- 2023StyleGAN-T: Unlocking the Power of GANs for Fast Large-Scale Text-to-Image Synthesiscited by 212
- 2021Projected GANs Converge Fastercited by 208
- 2021Diffusion-Based Voice Conversion with Fast Maximum Likelihood Sampling Schemecited by 100
- 2018StarGAN-VC: Non-parallel many-to-many voice conversion with star generative adversarial networkscited by 78
- 2021UnivNet: A Neural Vocoder with Multi-Resolution Spectrogram Discriminators for High-Fidelity Waveform Generationcited by 74
- 2020Multi-band MelGAN: Faster Waveform Generation for High-Quality Text-to-Speechcited by 63
- 2021StarGANv2-VC: A Diverse, Unsupervised, Non-parallel Framework for Natural-Sounding Voice Conversioncited by 58
- 2020Any-to-Many Voice Conversion with Location-Relative Sequence-to-Sequence Modelingcited by 42
- 2017Voice Conversion from Unaligned Corpora using Variational Autoencoding Wasserstein Generative Adversarial Networkscited by 38
- 2019StarGAN-VC2: Rethinking Conditional Methods for StarGAN-Based Voice Conversioncited by 37
- 2022iSTFTNet: Fast and Lightweight Mel-Spectrogram Vocoder Incorporating Inverse Short-Time Fourier Transformcited by 32
- 2021MaskCycleGAN-VC: Learning Non-parallel Voice Conversion with Filling in Framescited by 18
- 2017Statistical Parametric Speech Synthesis Incorporating Generative Adversarial Networkscited by 16
- 2020VoiceGrad: Non-Parallel Any-to-Many Voice Conversion with Annealed Langevin Dynamicscited by 14
- 2018Generative adversarial network-based approach to signal reconstruction from magnitude spectrogramscited by 8
- 2024FastVoiceGrad: One-step Diffusion-Based Voice Conversion with Adversarial Conditional Diffusion Distillationcited by 7
- 2023Wave-U-Net Discriminator: Fast and Lightweight Discriminator for Generative Adversarial Network-Based Speech Synthesiscited by 4
- 2023iSTFTNet2: Faster and More Lightweight iSTFT-Based Neural Vocoder Using 1D-2D CNNcited by 4
Built on by
Citing papers per year: 2026: 2.
2 papers.