Stabilizing label assignment for speech separation by self-supervised pre-training
Journal
Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Journal Volume
3
Pages
2303-2307
Date Issued
2021
Author(s)
Abstract
Speech separation has been well developed, with the very successful permutation invariant training (PIT) approach, although the frequent label assignment switching happening during PIT training remains to be a problem when better convergence speed and achievable performance are desired. In this paper, we propose to perform self-supervised pre-training to stabilize the label assignment in training the speech separation model. Experiments over several types of self-supervised approaches, several typical speech separation models and two different datasets showed that very good improvements are achievable if a proper self-supervised approach is chosen. Copyright ? 2021 ISCA.
Subjects
Label Permutation Switch
Self-supervised Pre-train
Speech Enhancement
Speech Separation
Separation
Source separation
Speech analysis
Speech communication
Achievable performance
Convergence speed
Label permutation switch
Pre-training
Self-supervised pre-train
Separation model
Speech separation
Speed performance
Speech enhancement
Type
conference paper