TF-CorrNet: Leveraging Spatial Correlation for Continuous Speech Separation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915900479766528 |
|---|---|
| author | Shin, Ui-Hyeop Ku, Bon Hyeok Park, Hyung-Min |
| author_facet | Shin, Ui-Hyeop Ku, Bon Hyeok Park, Hyung-Min |
| contents | In general, multi-channel source separation has utilized inter-microphone phase differences (IPDs) concatenated with magnitude information in time-frequency domain, or real and imaginary components stacked along the channel axis. However, the spatial information of a sound source is fundamentally contained in the differences between microphones, specifically in the correlation between them, while the power of each microphone also provides valuable information about the source spectrum, which is why the magnitude is also included. Therefore, we propose a network that directly leverages a correlation input with phase transform (PHAT)-beta to estimate the separation filter. In addition, the proposed TF-CorrNet processes the features alternately across time and frequency axes as a dual-path strategy in terms of spatial information. Furthermore, we add a spectral module to model source-related direct time-frequency patterns for improved speech separation. Experimental results demonstrate that the proposed TF-CorrNet effectively separates the speech sounds, showing high performance with a low computational cost in the LibriCSS dataset. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_16481 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TF-CorrNet: Leveraging Spatial Correlation for Continuous Speech Separation Shin, Ui-Hyeop Ku, Bon Hyeok Park, Hyung-Min Audio and Speech Processing Sound In general, multi-channel source separation has utilized inter-microphone phase differences (IPDs) concatenated with magnitude information in time-frequency domain, or real and imaginary components stacked along the channel axis. However, the spatial information of a sound source is fundamentally contained in the differences between microphones, specifically in the correlation between them, while the power of each microphone also provides valuable information about the source spectrum, which is why the magnitude is also included. Therefore, we propose a network that directly leverages a correlation input with phase transform (PHAT)-beta to estimate the separation filter. In addition, the proposed TF-CorrNet processes the features alternately across time and frequency axes as a dual-path strategy in terms of spatial information. Furthermore, we add a spectral module to model source-related direct time-frequency patterns for improved speech separation. Experimental results demonstrate that the proposed TF-CorrNet effectively separates the speech sounds, showing high performance with a low computational cost in the LibriCSS dataset. |
| title | TF-CorrNet: Leveraging Spatial Correlation for Continuous Speech Separation |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2509.16481 |