Self-Distillation Prototypes Network: Learning Robust Speaker Representations Without Supervision

doi:doi:10.57702/8qa3kqao

Self-Distillation Prototypes Network: Learning Robust Speaker Representations Without Supervision

Training speaker-discriminative and robust speaker verification systems without explicit speaker labels remains a persisting challenge. In this paper, we propose a new self-supervised speaker verification approach, Self-Distillation Prototypes Network (SDPN), which effectively facilitates self-supervised speaker representation learning.

BibTex:

@dataset{Yafeng_Chen_and_Siqi_Zheng_and_Hui_Wang_and_Luyao_Cheng_and_Qian_Chen_and_Shiliang_Zhang_and_Wen_Wang_2024,
    abstract = {Training speaker-discriminative and robust speaker verification systems without explicit speaker labels remains a persisting challenge. In this paper, we propose a new self-supervised speaker verification approach, Self-Distillation Prototypes Network (SDPN), which effectively facilitates self-supervised speaker representation learning.},
    author = {Yafeng Chen and Siqi Zheng and Hui Wang and Luyao Cheng and Qian Chen and Shiliang Zhang and Wen Wang},
    doi = {10.57702/8qa3kqao},
    institution = {No Organization},
    keyword = {'Self-Supervised Learning', 'Speaker Verification', 'VoxCeleb'},
    month = {dec},
    publisher = {TIB},
    title = {Self-Distillation Prototypes Network: Learning Robust Speaker Representations Without Supervision},
    url = {https://service.tib.eu/ldmservice/dataset/self-distillation-prototypes-network--learning-robust-speaker-representations-without-supervision},
    year = {2024}
}