6 papers
Environmental Sound Deepfake Detection Using Deep-Learning Framework
Khoi Vu, Dat Tran, Khanh Do +8
In this paper, we propose a deep-learning framework for Environmental Sound Deepfake Detection (ESDD) - the task of identifying whether the sound scene and sound event in an input…
A General Model for Deepfake Speech Detection: Diverse Bonafide Resources or Diverse AI-Based Generators
Lam Pham, Khoi Vu, Dat Tran +4
In this paper, we analyze two main factors of Bonafide Resource (BR) or AI-based Generator (AG) which affect the performance and the generality of a Deepfake Speech Detection (DSD)…
Diversity-Aware Reverse Kullback-Leibler Divergence for Large Language Model Distillation
Hoang-Chau Luong, Dat Ba Tran, Lingwei Chen
Reverse Kullback-Leibler (RKL) divergence has recently emerged as the preferred objective for large language model (LLM) distillation, consistently outperforming forward KL (FKL),…
Understanding SAM's Robustness to Noisy Labels through Gradient Down-weighting
Hoang-Chau Luong, Quang-Thuc Nguyen, Dat Ba Tran +1
Sharpness-Aware Minimization (SAM) was introduced to improve generalization by seeking flat minima, yet it also exhibits robustness to label noise, a phenomenon that remains only p…
Aud-Sur: An Audio Analyzer Assistant for Audio Surveillance Applications
Phat Lam, Lam Pham, Dat Tran +5
In this paper, we present an audio analyzer assistant tool designed for a wide range of audio-based surveillance applications (This work is a part of our DEFAME FAKES and EUCINF pr…
DIN-CTS: Low-Complexity Depthwise-Inception Neural Network with Contrastive Training Strategy for Deepfake Speech Detection
Lam Pham, Dat Tran, Phat Lam +5
In this paper, we propose a deep neural network approach for deepfake speech detection (DSD) based on a lowcomplexity Depthwise-Inception Network (DIN) trained with a contrastive t…