Hubert-Derived SSL Features and ECAPA-TDNN Matching for Robust Audio Deepfake Detection

Gul Tahaoglu. Hubert-Derived SSL Features and ECAPA-TDNN Matching for Robust Audio Deepfake Detection. In 35th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2025, Istanbul, Turkey, August 31 - Sept. 3, 2025. pages 1-6, IEEE, 2025. [doi]

@inproceedings{Tahaoglu25,
  title = {Hubert-Derived SSL Features and ECAPA-TDNN Matching for Robust Audio Deepfake Detection},
  author = {Gul Tahaoglu},
  year = {2025},
  doi = {10.1109/MLSP62443.2025.11204297},
  url = {https://doi.org/10.1109/MLSP62443.2025.11204297},
  researchr = {https://researchr.org/publication/Tahaoglu25},
  cites = {0},
  citedby = {0},
  pages = {1-6},
  booktitle = {35th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2025, Istanbul, Turkey, August 31 - Sept. 3, 2025},
  publisher = {IEEE},
  isbn = {979-8-3315-7029-3},
}