Abstract
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 perceptual quality of enhanced speech. The proposed model generalizes well to new speakers, new speech content, and new environments. It significantly outperforms state-of-the-art baseline methods in both objective and subjective experiments.
Original language | English (US) |
---|---|
Pages (from-to) | 4506-4510 |
Number of pages | 5 |
Journal | Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH |
Volume | 2020-October |
DOIs | |
State | Published - 2020 |
Event | 21st Annual Conference of the International Speech Communication Association, INTERSPEECH 2020 - Shanghai, China Duration: Oct 25 2020 → Oct 29 2020 |
All Science Journal Classification (ASJC) codes
- Language and Linguistics
- Human-Computer Interaction
- Signal Processing
- Software
- Modeling and Simulation
Keywords
- Deep features
- Denoising
- Dereverberation
- Generative adversarial networks
- Speech enhancement