Publication detail

Progressive contrastive learning for self-supervised text-independent speaker verification

PENG, J. ZHANG, C. ČERNOCKÝ, J. YU, D.

Original Title

Progressive contrastive learning for self-supervised text-independent speaker verification

Type

article in a collection out of WoS and Scopus

Language

English

Original Abstract

Self-supervised speaker representation learning has drawn attention extensively in recent years. Most of the work is based on the iterative clustering-classification learning framework, and the performance is sensitive to the pre-defined number of clusters. However, the cluster number is hard to estimate when dealing with large-scale unlabeled data. In this paper, we propose a progressive contrastive learning (PCL) algorithm to dynamically estimate the cluster number at each step based on the statistical characteristics of the data itself, and the estimated number will progressively approach the ground-truth speaker number with the increasing of step. Specifically, we first update the data queue by current augmented samples. Then, eigendecomposition is introduced to estimate the number of speakers in the updated data queue. Finally, we assign the queued data into the estimated cluster centroid and construct a contrastive loss, which encourages the speaker representation to be closer to its cluster centroid and away from others. Experimental results on VoxCeleb1 demonstrate the effectiveness of our proposed PCL compared with existing self-supervised approaches.

Keywords

self-supervised, text-independent, speaker, verification

Authors

PENG, J.; ZHANG, C.; ČERNOCKÝ, J.; YU, D.

Released

28. 6. 2022

Publisher

International Speech Communication Association

Location

Beijing

Pages from

17

Pages to

24

Pages count

8

URL

https://www.isca-speech.org/archive/pdfs/odyssey_2022/peng22_odyssey.pdf

BibTex

@inproceedings{BUT179661,
  author="Junyi {Peng} and Chunlei {Zhang} and Jan {Černocký} and Dong {Yu}",
  title="Progressive contrastive learning for self-supervised text-independent speaker verification",
  booktitle="Proceedings of The Speaker and Language Recognition Workshop (Odyssey 2022)",
  year="2022",
  pages="17--24",
  publisher="International Speech Communication Association",
  address="Beijing",
  doi="10.21437/Odyssey.2022-3",
  url="https://www.isca-speech.org/archive/pdfs/odyssey_2022/peng22_odyssey.pdf"
}