English

Fre-GAN: Adversarial Frequency-consistent Audio Synthesis

Audio and Speech Processing 2021-06-15 v2 Machine Learning

Abstract

Although recent works on neural vocoder have improved the quality of synthesized audio, there still exists a gap between generated and ground-truth audio in frequency space. This difference leads to spectral artifacts such as hissing noise or reverberation, and thus degrades the sample quality. In this paper, we propose Fre-GAN which achieves frequency-consistent audio synthesis with highly improved generation quality. Specifically, we first present resolution-connected generator and resolution-wise discriminators, which help learn various scales of spectral distributions over multiple frequency bands. Additionally, to reproduce high-frequency components accurately, we leverage discrete wavelet transform in the discriminators. From our experiments, Fre-GAN achieves high-fidelity waveform generation with a gap of only 0.03 MOS compared to ground-truth audio while outperforming standard models in quality.

Keywords

Cite

@article{arxiv.2106.02297,
  title  = {Fre-GAN: Adversarial Frequency-consistent Audio Synthesis},
  author = {Ji-Hoon Kim and Sang-Hoon Lee and Ji-Hyun Lee and Seong-Whan Lee},
  journal= {arXiv preprint arXiv:2106.02297},
  year   = {2021}
}

Comments

Accepted paper in Interspeech 2021

R2 v1 2026-06-24T02:49:41.035Z