Abstract
Numerous methods exist for sound source separation, leveraging either classical signal processing or deep learning approaches. While deep-learning-based models often outperform conventional methods, they require large training datasets and struggle to generalize to new settings. To address this, we propose Deep-Simplex, a deep prior-based method that reconstructs the probability simplex of speaker activity over time. This global activity probability guides the estimation of a local mask per frequency, identifying the dominant speaker in each time-frequency (TF) bin. We then use this mask for both spatial and spectral separation. Experimental results demonstrate that Deep-Simplex outperforms competing baselines in different reverberation conditions.
| Original language | English |
|---|---|
| Pages (from-to) | 1463-1467 |
| Number of pages | 5 |
| Journal | Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH |
| DOIs | |
| State | Published - 2025 |
| Externally published | Yes |
| Event | 26th Interspeech Conference 2025 - Rotterdam, Netherlands Duration: 17 Aug 2025 → 21 Aug 2025 |
Bibliographical note
Publisher Copyright:© 2025 International Speech Communication Association. All rights reserved.
Keywords
- deep prior
- simplex
- source separation
Fingerprint
Dive into the research topics of 'Deep-Simplex Multichannel Speech Separation'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver