Comprehensive Layer-wise Analysis of SSL Models for Audio Deepfake Detection
Open paper
Benchmark
Model
Rank
Results
voice-anti-spoofing-on-asvspoof-2019-la
Wav2Vec2-Large-12+AASIST
#1
EER: 0.21
voice-anti-spoofing-on-asvspoof-2019-la
Wav2Vec2-Small-6+FFN
#2
EER: 0.48
Rank counts only results with a code link.