Analyzing the impact of splicing artifacts in partially fake speech signals
Speech deepfake detection has recently gained significant attention within the multimedia
forensics community. Related issues have also been explored, such as the identification of …
forensics community. Related issues have also been explored, such as the identification of …
Toward Robust Real-World Audio Deepfake Detection: Closing the Explainability Gap
The rapid proliferation of AI-manipulated or generated audio deepfakes poses serious
challenges to media integrity and election security. Current AI-driven detection solutions lack …
challenges to media integrity and election security. Current AI-driven detection solutions lack …
BUT Systems and Analyses for the ASVspoof 5 Challenge
This paper describes the BUT submitted systems for the ASVspoof 5 challenge, along with
analyses. For the conventional deepfake detection task, we use ResNet18 and self …
analyses. For the conventional deepfake detection task, we use ResNet18 and self …
USTC-KXDIGIT System Description for ASVspoof5 Challenge
Y Chen, H Wu, N Jiang, X **a, Q Gu, Y Hao… - arxiv preprint arxiv …, 2024 - arxiv.org
This paper describes the USTC-KXDIGIT system submitted to the ASVspoof5 Challenge for
Track 1 (speech deepfake detection) and Track 2 (spoofing-robust automatic speaker …
Track 1 (speech deepfake detection) and Track 2 (spoofing-robust automatic speaker …
Temporal Variability and Multi-Viewed Self-Supervised Representations to Tackle the ASVspoof5 Deepfake Challenge
ASVspoof5, the fifth edition of the ASVspoof series, is one of the largest global audio security
challenges. It aims to advance the development of countermeasure (CM) to discriminate …
challenges. It aims to advance the development of countermeasure (CM) to discriminate …
SZU-AFS Antispoofing System for the ASVspoof 5 Challenge
This paper presents the SZU-AFS anti-spoofing system, designed for Track 1 of the
ASVspoof 5 Challenge under open conditions. The system is built with four stages: selecting …
ASVspoof 5 Challenge under open conditions. The system is built with four stages: selecting …
ASASVIcomtech: the Vicomtech-UGR speech deepfake detection and SASV systems for the ASVspoof5 Challenge
This paper presents the work carried out by the ASASVIcomtech team, made up of
researchers from Vicomtech and University of Granada, for the ASVspoof5 Challenge. The …
researchers from Vicomtech and University of Granada, for the ASVspoof5 Challenge. The …
LlamaPartialSpoof: An LLM-Driven Fake Speech Dataset Simulating Disinformation Generation
Previous fake speech datasets were constructed from a defender's perspective to develop
countermeasure (CM) systems without considering diverse motivations of attackers. To …
countermeasure (CM) systems without considering diverse motivations of attackers. To …
Exploring Generalization To Unseen Audio Data For Spoofing: Insights From SSL Models
Deep learning-based speech synthesis has significantly improved realistic audio deepfakes.
Despite advanced techniques such as self-supervised learning (SSL) and datasets, current …
Despite advanced techniques such as self-supervised learning (SSL) and datasets, current …
Data augmentations for audio deepfake detection for the ASVspoof5 closed condition
R Duroselle, O Boeffard, A Courtois, H Nourtel… - ASVspoof Workshop …, 2024 - hal.science
This paper describes the joint participation of Inria Défense et Sécurité and Storyzy to the
ASVspoof5 challenge. We participated in the closed conditions of the audio deepfake …
ASVspoof5 challenge. We participated in the closed conditions of the audio deepfake …