UniPET-SPK: A Unified Framework for Parameter-Efficient Tuning of Pre-trained Speech Models for Robust Speaker Verification

Open paper
Benchmark Model Rank Results
speaker-verification-on-voxceleb1-e-cleanUniPET-SPK (WavLM Base+ + TDNN)–EER: 1.74minDCF (P_target=0.01): 0.138
speaker-verification-on-voxceleb1-h-cleanUniPET-SPK (WavLM Base+ + TDNN)–EER: 3.41minDCF (P_target=0.01): 0.206
speaker-verification-on-voxceleb1-o-cleanUniPET-SPK (WavLM Base+ + TDNN)–EER: 1.59minDCF (P_target=0.01): 0.130

Rank counts only results with a code link.