JenGAN: Stacked Shifted Filters in GAN-Based Speech Synthesis
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909220276797440 |
|---|---|
| author | Cho, Hyunjae Lee, Junhyeok Jung, Wonbin |
| author_facet | Cho, Hyunjae Lee, Junhyeok Jung, Wonbin |
| contents | Non-autoregressive GAN-based neural vocoders are widely used due to their fast inference speed and high perceptual quality. However, they often suffer from audible artifacts such as tonal artifacts in their generated results. Therefore, we propose JenGAN, a new training strategy that involves stacking shifted low-pass filters to ensure the shift-equivariant property. This method helps prevent aliasing and reduce artifacts while preserving the model structure used during inference. In our experimental evaluation, JenGAN consistently enhances the performance of vocoder models, yielding significantly superior scores across the majority of evaluation metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_06111 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | JenGAN: Stacked Shifted Filters in GAN-Based Speech Synthesis Cho, Hyunjae Lee, Junhyeok Jung, Wonbin Audio and Speech Processing Artificial Intelligence Sound Signal Processing Non-autoregressive GAN-based neural vocoders are widely used due to their fast inference speed and high perceptual quality. However, they often suffer from audible artifacts such as tonal artifacts in their generated results. Therefore, we propose JenGAN, a new training strategy that involves stacking shifted low-pass filters to ensure the shift-equivariant property. This method helps prevent aliasing and reduce artifacts while preserving the model structure used during inference. In our experimental evaluation, JenGAN consistently enhances the performance of vocoder models, yielding significantly superior scores across the majority of evaluation metrics. |
| title | JenGAN: Stacked Shifted Filters in GAN-Based Speech Synthesis |
| topic | Audio and Speech Processing Artificial Intelligence Sound Signal Processing |
| url | https://arxiv.org/abs/2406.06111 |