mlsanityData to 14 Oct 2026

Unified Source-Filter GAN with Harmonic-plus-Noise Source Excitation Generation

arXiv 2205.06053 ↗First posted 12 May 2022cs.SD cs.LG eess.AScited by 5

Not on the map. Why?

Abstract (arXiv)

This paper introduces a unified source-filter network with a harmonic-plus-noise source excitation generation mechanism. In our previous work, we proposed unified Source-Filter GAN (uSFGAN) for developing a high-fidelity neural vocoder with flexible voice controllability using a unified source-filter neural network architecture. However, the capability of uSFGAN to model the aperiodic source excitation signal is insufficient, and there is still a gap in sound quality between the natural and generated speech. To improve the source excitation modeling and generated sound quality, a new source excitation generation network separately generating periodic and aperiodic components is proposed. The advanced adversarial training procedure of HiFiGAN is also adopted to replace that of Parallel WaveGAN used in the original uSFGAN. Both objective and subjective evaluation results show that the modified uSFGAN significantly improves the sound quality of the basic uSFGAN while maintaining the voice controllability.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 15 of its 22 references that are in the map. First posted on arXiv 12 May 2022.

Builds on

15 of 22 references.

References · 15

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2022Generative Adversarial Networkscited by 5,627
  3. 2016WaveNet: A Generative Model for Raw Audiocited by 1,722
  4. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,154
  5. 2019On the Variance of the Adaptive Learning Rate and Beyondcited by 772
  6. 2019Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogramcited by 340
  7. 2018Efficient Neural Audio Synthesiscited by 219
  8. 2018LPCNet: Improving Neural Speech Synthesis Through Linear Predictioncited by 110
  9. 2018Neural source-filter-based waveform model for statistical parametric speech synthesiscited by 22
  10. 2021GAN Vocoder: Multi-Resolution Discriminator Is All You Needcited by 16
  11. 2019Neural Harmonic-plus-Noise Waveform Model with Trainable Maximum Voice Frequency for Text-to-Speech Synthesiscited by 11
  12. 2020Quasi-Periodic Parallel WaveGAN: A Non-autoregressive Raw Waveform Generative Model with Pitch-dependent Dilated Convolution Neural Networkcited by 8
3 more
  1. 2021PeriodNet: A non-autoregressive waveform generation model with a structure separating periodic and aperiodic componentscited by 7
  2. 2020Quasi-Periodic WaveNet: An Autoregressive Raw Waveform Generative Model with Pitch-dependent Dilated Convolution Neural Networkcited by 6
  3. 2021Unified Source-Filter GAN: Unified Source-filter Network Based On Factorization of Quasi-Periodic Parallel WaveGANcited by 3

Built on by

Cited by year

Citing papers per year: 2022: 3; 2023: 2.

5 papers.

  1. 2023The Singing Voice Conversion Challenge 2023cited by 25
  2. 2022Source-Filter HiFi-GAN: Fast and Pitch Controllable High-Fidelity Neural Vocodercited by 11
  3. 2023HiFTNet: A Fast High-Quality Neural Vocoder with Harmonic-plus-Noise Filter and Inverse Short Time Fourier Transformcited by 11
  4. 2022Period VITS: Variational Inference with Explicit Pitch Modeling for End-to-end Emotional Speech Synthesiscited by 6
  5. 2022NNSVS: A Neural Network-Based Singing Voice Synthesis Toolkitcited by 3