Discriminative Speaker Representation Via Contrastive Learning With Class-aware Attention In Angular Space

Abstract

The challenges in applying contrastive learning to speaker verification (SV) are that the softmax-based contrastive loss lacks discriminative power and that the hard negative pairs can easily influence learning. To overcome the first challenge, we propose a contrastive learning SV framework incorporating an additive angular margin into the supervised contrastive loss in which the margin improves the speaker representation's discrimination ability. For the second challenge, we introduce a class-aware attention mechanism through which hard negative samples contribute less significantly to the supervised contrastive loss. We also employed gradient-based multi-objective optimization to balance the classification and contrastive loss. Experimental results on CN-Celeb and Voxceleb1 show that this new learning objective can cause the encoder to find an embedding space that exhibits great speaker discrimination across languages.

Stats

• citations13
• S2 citations—
• github stars0
• HF likes0
• heat score8.60
• arxiv keyli2022discriminative

Related papers

• Speaker Representation Learning Via Contrastive Loss With Maximal Speaker Separability (2022)10.68
• A Study On Angular Based Embedding Learning For Text-independent Speaker Verification (2019)2.26
• Experimenting With Additive Margins For Contrastive Self-supervised Speaker Verification (2023)4.52
• Additive Margin In Contrastive Self-supervised Frameworks To Learn Discriminative Speaker Representations (2024)2.26
• Asymmetric Clean Segments-guided Self-supervised Learning For Robust Speaker Verification (2023)5.84
• Angular Softmax Loss For End-to-end Speaker Verification (2018)11.19
• Self-supervised Learning Of Audio Representations Using Angular Contrastive Loss (2022)5.84
• Self-supervised Text-independent Speaker Verification Using Prototypical Momentum Contrastive Learning (2020)12.93