ASVspoof-2021
Emerging10papers using it
2022first seen
The 'ASVspoof-2021' dataset is a benchmark used to evaluate the performance of automatic speaker verification systems against sophisticated spoofing attacks, specifically focusing on synthetic speech detection.
Papers using ASVspoof-2021 (10)
- Pushing the Performance of Synthetic Speech Detection with Kolmogorov-Arnold Networks and Self-Supervised Learning ModelsXLSR-Kanformer: A KAN-Intergrated model for Synthetic Speech DetectionAmplifying Artifacts with Speech Enhancement in Voice Anti-spoofingNes2Net: A Lightweight Nested Architecture for Foundation Model Driven Speech Anti-spoofingThe Vicomtech Audio Deepfake Detection System based on Wav2Vec2 for the
2022 ADD ChallengeImproved DeepFake Detection Using Whisper FeaturesCharacterizing the temporal dynamics of universal speech representations
for generalizable deepfake detectionCollaborative Watermarking for Adversarial Speech SynthesisTemporal-Channel Modeling in Multi-head Self-Attention for Synthetic
Speech DetectionMixture of Experts Fusion for Fake Audio Detection Using Frozen wav2vec
2.0