ECAPA2: A Hybrid Neural Network Architecture and Training Strategy for Robust Speaker Embeddings

Open paper
Benchmark Model Rank Results
speaker-verification-on-voxceleb1-e-cleanECAPA2–EER: 0.52minDCF (P_target=0.01): 0.058
speaker-verification-on-voxceleb1-h-cleanECAPA2–EER: 0.99minDCF (P_target=0.01): 0.098
speaker-verification-on-voxceleb1-o-cleanECAPA2–EER: 0.34minDCF (P_target=0.01): 0.029

Rank counts only results with a code link.