Bootstrap Equilibrium and Probabilistic Speaker Representation Learning for Self-Supervised Speaker Verification

In this paper, we propose self-supervised speaker representation learning strategies, which comprise of a bootstrap equilibrium speaker representation learning in the front-end and an uncertainty-aware probabilistic speaker embedding training in the back-end. In the front-end stage, we learn the spe...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2021, Vol.9, p.167615-167627
Main Authors: Mun, Sung Hwan, Han, Min Hyun, Lee, Dongjune, Kim, Jihwan, Kim, Nam Soo
Format: Article
Language:English
Subjects:
bootstrap representation learning
Entropy
Equilibrium
Learning
Maximum likelihood estimation
Probabilistic logic
probabilistic speaker embedding
Regularization
Representation learning
Representations
self-supervised learning
Speaker verification
Task analysis
Training
Uncertainty
Verification
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we propose self-supervised speaker representation learning strategies, which comprise of a bootstrap equilibrium speaker representation learning in the front-end and an uncertainty-aware probabilistic speaker embedding training in the back-end. In the front-end stage, we learn the speaker representations via the bootstrap training scheme with the uniformity regularization term. In the back-end stage, the probabilistic speaker embeddings are estimated by maximizing the mutual likelihood score between the speech samples belonging to the same speaker, which provide not only speaker representations but also data uncertainty. Experimental results show that the proposed bootstrap equilibrium training strategy can effectively help learn the speaker representations and outperforms the conventional methods based on contrastive learning. Also, we demonstrate that the integrated two-stage framework further improves the speaker verification performance on the VoxCeleb1 test set in terms of EER and MinDCF.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2021.3137190