Unrestricted Global Phase Bias-Aware Single-channel Speech Enhancement with Conformer-based Metric GAN
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_ | 1866911902984044544 |
|---|---|
| author | Zhang, Shiqi Qiu, Zheng Takeuchi, Daiki Harada, Noboru Makino, Shoji |
| author_facet | Zhang, Shiqi Qiu, Zheng Takeuchi, Daiki Harada, Noboru Makino, Shoji |
| contents | With the rapid development of neural networks in recent years, the ability of various networks to enhance the magnitude spectrum of noisy speech in the single-channel speech enhancement domain has become exceptionally outstanding. However, enhancing the phase spectrum using neural networks is often ineffective, which remains a challenging problem. In this paper, we found that the human ear cannot sensitively perceive the difference between a precise phase spectrum and a biased phase (BP) spectrum. Therefore, we propose an optimization method of phase reconstruction, allowing freedom on the global-phase bias instead of reconstructing the precise phase spectrum. We applied it to a Conformer-based Metric Generative Adversarial Networks (CMGAN) baseline model, which relaxes the existing constraints of precise phase and gives the neural network a broader learning space. Results show that this method achieves a new state-of-the-art performance without incurring additional computational overhead. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_08252 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Unrestricted Global Phase Bias-Aware Single-channel Speech Enhancement with Conformer-based Metric GAN Zhang, Shiqi Qiu, Zheng Takeuchi, Daiki Harada, Noboru Makino, Shoji Audio and Speech Processing Sound With the rapid development of neural networks in recent years, the ability of various networks to enhance the magnitude spectrum of noisy speech in the single-channel speech enhancement domain has become exceptionally outstanding. However, enhancing the phase spectrum using neural networks is often ineffective, which remains a challenging problem. In this paper, we found that the human ear cannot sensitively perceive the difference between a precise phase spectrum and a biased phase (BP) spectrum. Therefore, we propose an optimization method of phase reconstruction, allowing freedom on the global-phase bias instead of reconstructing the precise phase spectrum. We applied it to a Conformer-based Metric Generative Adversarial Networks (CMGAN) baseline model, which relaxes the existing constraints of precise phase and gives the neural network a broader learning space. Results show that this method achieves a new state-of-the-art performance without incurring additional computational overhead. |
| title | Unrestricted Global Phase Bias-Aware Single-channel Speech Enhancement with Conformer-based Metric GAN |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2402.08252 |