Please use this identifier to cite or link to this item:
http://bura.brunel.ac.uk/handle/2438/29080
Title: | mdctGAN: Taming transformer-based GAN for speech super-resolution with Modified DCT spectra |
Authors: | Shuai, C Shi, C Gan, L Liu, H |
Keywords: | speech super-resolution;phase information;GAN |
Issue Date: | 20-Aug-2023 |
Publisher: | ISCA |
Citation: | Shuai, C. et al. (2023) 'mdctGAN: Taming transformer-based GAN for speech super-resolution with Modified DCT spectra', Proceedings of INTERSPEECH 2023, Dublin, Ireland, 20-24 August, pp. 5112 - 5116. doi: 10.21437/interspeech.2023-113. |
Abstract: | Speech super-resolution (SSR) aims to recover a high resolution (HR) speech from its corresponding low resolution (LR) counterpart. Recent SSR methods focus more on the reconstruction of the magnitude spectrogram, ignoring the importance of phase reconstruction, thereby limiting the recovery quality. To address this issue, we propose mdctGAN, a novel SSR framework based on modified discrete cosine transform (MDCT). By adversarial learning in the MDCT domain, our method reconstructs HR speeches in a phase-aware manner without vocoders or additional post-processing. Furthermore, by learning frequency consistent features with self-attentive mechanism, mdctGAN guarantees high quality speech reconstruction. For VCTK corpus dataset, the experiment results show that our model produces natural auditory quality with high MOS and PESQ scores. It also achieves the state-of-the-art log-spectral-distance (LSD) performance on 48 kHz target resolution from various input rates. Code is available from https://github.com/neoncloud/mdctGAN |
Description: | Annual Conference of the International Speech Communication Association |
URI: | https://bura.brunel.ac.uk/handle/2438/29080 |
DOI: | https://doi.org/10.21437/interspeech.2023-113 |
Other Identifiers: | ORCiD: Lu Gan https://orcid.org/0000-0003-1056-7660 |
Appears in Collections: | Dept of Electronic and Electrical Engineering Research Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
FullText.pdf | Copyright © 2023 The Author(s). This work is licensed under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). | 1.3 MB | Adobe PDF | View/Open |
This item is licensed under a Creative Commons License