• DOI:10.21437/interspeech.2020-2143
Corpus ID: 219558473

HiFi-GAN: High-Fidelity Denoising and Dereverberation Based on Speech Deep Features in Adversarial Networks

@inproceedings{Su2020HiFiGANHD,
  title={HiFi-GAN: High-Fidelity Denoising and Dereverberation Based on Speech Deep Features in Adversarial Networks},
  author={Jiaqi Su and Zeyu Jin and A. Finkelstein},
  booktitle={INTERSPEECH},
  year={2020}
}

• Jiaqi Su, Zeyu Jin, A. Finkelstein
• Published in INTERSPEECH 2020
• Engineering, Computer Science

Real-world audio recordings are often degraded by factors such as noise, reverberation, and equalization distortion. This paper introduces HiFi-GAN, a deep learning method to transform recorded speech to sound as though it had been recorded in a studio. We use an end-to-end feed-forward WaveNet architecture, trained with multi-scale adversarial discriminators in both the time domain and the time-frequency domain. It relies on the deep feature matching losses of the discriminators to improve the… Expand

View PDF on arXiv

Save to Library

Create Alert

Cite

Launch Research Feed