Self-supervised Speaker Verification Using Dynamic Loss-gate And Label Correction

Abstract

For self-supervised speaker verification, the quality of pseudo labels decides the upper bound of the system due to the massive unreliable labels. In this work, we propose dynamic loss-gate and label correction (DLG-LC) to alleviate the performance degradation caused by unreliable estimated labels. In DLG, we adopt Gaussian Mixture Model (GMM) to dynamically model the loss distribution and use the estimated GMM to distinguish the reliable and unreliable labels automatically. Besides, to better utilize the unreliable data instead of dropping them directly, we correct the unreliable label with model predictions. Moreover, we apply the negative-pairs-free DINO framework in our experiments for further improvement. Compared to the best-known speaker verification system with self-supervised learning, our proposed DLG-LC converges faster and achieves 11.45%, 18.35% and 15.16% relative improvement on Vox-O, Vox-E and Vox-H trials of Voxceleb1 evaluation dataset.

Stats

• citations26
• S2 citations—
• github stars0
• HF likes0
• heat score10.74
• arxiv keyhan2022self

Related papers

• Self-supervised Speaker Recognition With Loss-gated Learning (2021)16.93
• Self-supervised Learning With Cluster-aware-dino For High-performance Robust Speaker Verification (2023)0.00
• Self-supervised Learning Based Domain Adaptation For Robust Speaker Verification (2021)11.49
• Pushing The Limits Of Self-supervised Speaker Verification Using Regularized Distillation Framework (2022)17.00
• Asymmetric Clean Segments-guided Self-supervised Learning For Robust Speaker Verification (2023)5.84
• Curriculum Learning For Self-supervised Speaker Verification (2022)8.09
• Experimenting With Additive Margins For Contrastive Self-supervised Speaker Verification (2023)4.52
• Self-supervised Speaker Verification With Simple Siamese Network And Self-supervised Regularization (2021)10.85