ASVspoof 2019
Emerging20papers using it
2023first seen
ASVspoof 2019 is a dataset used to evaluate the effectiveness of systems in detecting speech deepfakes and spoofing attacks.
Papers using ASVspoof 2019 (19)
- Heterogeneity Over Homogeneity: Investigating Multilingual Speech Pre-trained Models For Detecting Audio DeepfakeCharacterizing The Temporal Dynamics Of Universal Speech Representations For Generalizable Deepfake DetectionFusion of Modulation Spectrogram and SSL with Multi-head Attention for Fake Speech DetectionQuantizer-Aware Hierarchical Neural Codec Modeling for Speech Deepfake DetectionAmplifying Artifacts with Speech Enhancement in Voice Anti-spoofingATMM-SAGA: Alternating Training for Multi-Module with Score-Aware Gated Attention SASV systemDSVAE: Interpretable Disentangled Representation For Synthetic Speech DetectionCompression Robust Synthetic Speech Detection Using Patched Spectrogram TransformerInvestigating Prosodic Signatures Via Speech Pre-trained Models For Audio Deepfake Source AttributionMixture Of Experts Fusion For Fake Audio Detection Using Frozen Wav2vec 2.0DSVAE: Interpretable Disentangled Representation for Synthetic Speech
DetectionDeepfake Audio Detection Using Spectrogram-based Feature and Ensemble of
Deep Learning ModelsAudio Deepfake Detection with Self-Supervised WavLM and Multi-Fusion
Attentive ClassifierCompression Robust Synthetic Speech Detection Using Patched Spectrogram
TransformerCharacterizing the temporal dynamics of universal speech representations
for generalizable deepfake detectionAdvanced Signal Analysis in Detecting Replay Attacks for Automatic Speaker Verification SystemsHeterogeneity over Homogeneity: Investigating Multilingual Speech
Pre-Trained Models for Detecting Audio DeepfakeMixture of Experts Fusion for Fake Audio Detection Using Frozen wav2vec
2.0DiffSSD: A Diffusion-Based Dataset For Speech Forensics