# Parallel waveform synthesis based on generative adversarial networks with voicing-aware conditional discriminators

arXiv: https://arxiv.org/abs/2010.14151 · HTML: https://mlsanity.com/paper/2010.14151/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Oct 2020.

## Built on by (most cited first, 7 of 7)

- [Phrase break prediction with bidirectional encoder representations in Japanese text-to-speech synthesis](https://mlsanity.com/paper/2104.12395/) (2021)
- [Modeling and Driving Human Body Soundfields through Acoustic Primitives](https://mlsanity.com/paper/2407.13083/) (2024)
- [Sounding Bodies: Modeling 3D Spatial Sound of Humans Using Body Pose and Audio](https://mlsanity.com/paper/2311.06285/) (2023)
- [Learning to Highlight Audio by Watching Movies](https://mlsanity.com/paper/2505.12154/) (2025)
- [Improve GAN-based Neural Vocoder using Pointwise Relativistic LeastSquare GAN](https://mlsanity.com/paper/2103.14245/) (2021)
- [MSR-NV: Neural Vocoder Using Multiple Sampling Rates](https://mlsanity.com/paper/2109.13714/) (2021)
- [WOLONet: Wave Outlooker for Efficient and High Fidelity Speech Synthesis](https://mlsanity.com/paper/2206.09920/) (2022)

## Abstract (arXiv)

This paper proposes voicing-aware conditional discriminators for Parallel WaveGAN-based waveform synthesis systems. In this framework, we adopt a projection-based conditioning method that can significantly improve the discriminator's performance. Furthermore, the conventional discriminator is separated into two waveform discriminators for modeling voiced and unvoiced speech. As each discriminator learns the distinctive characteristics of the harmonic and noise components, respectively, the adversarial training process becomes more efficient, allowing the generator to produce more realistic speech waveforms. Subjective test results demonstrate the superiority of the proposed method over the conventional Parallel WaveGAN and WaveNet systems. In particular, our speaker-independently trained model within a FastSpeech 2 based text-to-speech framework achieves the mean opinion scores of 4.20, 4.18, 4.21, and 4.31 for four Japanese speakers, respectively.