Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.01130 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911188500086784 |
|---|---|
| author | Wang, Tingting Wang, Tianrui Ge, Meng Zhang, Qiquan Shao, Xi |
| author_facet | Wang, Tingting Wang, Tianrui Ge, Meng Zhang, Qiquan Shao, Xi |
| contents | The Graph Fourier Transform (GFT) has recently demonstrated promising results in speech enhancement. However, existing GFT-based speech enhancement approaches often employ fixed graph topologies to build the graph Fourier basis, whose the representation lacks the adaptively and flexibility. In addition, they suffer from the numerical errors and instability introduced by matrix inversion in GFT based on both Singular Value Decomposition (GFT-SVD) and Eigen Vector Decomposition (GFT-EVD). Motivated by these limitations, this paper propose a simple yet effective learnable GFT-SVD framework for speech enhancement. Specifically, we leverage graph shift operators to construct a learnable graph topology and define a learnable graph Fourier basis by the singular value matrices using 1-D convolution (Conv-1D) neural layer. This eliminates the need for matrix inversion, thereby avoiding the associated numerical errors and stability problem. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_01130 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning Time-Graph Frequency Representation for Monaural Speech Enhancement Wang, Tingting Wang, Tianrui Ge, Meng Zhang, Qiquan Shao, Xi Audio and Speech Processing The Graph Fourier Transform (GFT) has recently demonstrated promising results in speech enhancement. However, existing GFT-based speech enhancement approaches often employ fixed graph topologies to build the graph Fourier basis, whose the representation lacks the adaptively and flexibility. In addition, they suffer from the numerical errors and instability introduced by matrix inversion in GFT based on both Singular Value Decomposition (GFT-SVD) and Eigen Vector Decomposition (GFT-EVD). Motivated by these limitations, this paper propose a simple yet effective learnable GFT-SVD framework for speech enhancement. Specifically, we leverage graph shift operators to construct a learnable graph topology and define a learnable graph Fourier basis by the singular value matrices using 1-D convolution (Conv-1D) neural layer. This eliminates the need for matrix inversion, thereby avoiding the associated numerical errors and stability problem. |
| title | Learning Time-Graph Frequency Representation for Monaural Speech Enhancement |
| topic | Audio and Speech Processing |
| url | https://arxiv.org/abs/2510.01130 |