ASTDT: an Interpretable Adaptive Spectro-Temporal Diffusion Transformer for audio deepfake detection

Abstract Advances in audio synthesis techniques have led to the creation of highly realistic audio deepfakes, posing growing threats to digital integrity and public trust. These synthetic manipulations mimic natural speech with high fidelity, making detection increasingly challenging and fueling the...

Ausführliche Beschreibung

Bibliografische Detailangaben

Hauptverfasser:	Taiba Maijd Wani, Syed Asif Ahmad Qadri, Arselan Ashraf, Irene Amerini
Format:	Artigo
Sprache:	Inglês
Veröffentlicht:	SpringerOpen 2025-10-01
Schriftenreihe:	EURASIP Journal on Information Security
Schlagworte:	Audio deepfake detection Adaptive Spectro-Temporal Diffusion Transformer (ASTDT) Spectrogram analysis Dual-modal attention fusion Interpretability
Online-Zugang:	https://doi.org/10.1186/s13635-025-00217-3
Tags:	Tag hinzufügen Keine Tags, Fügen Sie das erste Tag hinzu!