SIGNL: A label-efficient audio deepfake detection system via spectral-temporal graph non-contrastive learning.
Saved in:
| Title: | SIGNL: A label-efficient audio deepfake detection system via spectral-temporal graph non-contrastive learning. |
|---|---|
| Authors: | Febrinanto, Falih Gozi1,2, f.febrinanto@federation.edu.au, Moore, Kristen2, kristen.moore@data61.csiro.au, Thapa, Chandra2, chandra.thapa@data61.csiro.au, Ma, Jiangang1, j.ma@federation.edu.au, Saikrishna, Vidya1, v.saikrishna@federation.edu.au |
| Source: | Expert Systems with Applications; Apr2026, Vol. 307, pN.PAG-N.PAG, 1p |
| Database: | Applied Science & Technology Source |
| ISSN: | 09574174 |
|---|---|
| DOI: | 10.1016/j.eswa.2025.131056 |
