Teacher-Student MixIT for Unsupervised and Semi-supervised Speech Separation
Abstract
In this paper, we introduce a novel semi-supervised learning framework for end-to-end speech separation. The proposed method first uses mixtures of unseparated sources and the mixture invariant training (MixIT) criterion to train a teacher model. The teacher model then estimates separated sources that are used to train a student model with standard permutation invariant training (PIT). The student model can be fine-tuned with supervised data, i.e., paired artificial mixtures and clean speech sources, and further improved via model distillation. Experiments with single and multi channel mixtures show that the teacher-student training resolves the over-separation problem observed in the original MixIT method. Further, the semisupervised performance is comparable to a fully-supervised separation system trained using ten times the amount of supervised data.
Keywords
Cite
@article{arxiv.2106.07843,
title = {Teacher-Student MixIT for Unsupervised and Semi-supervised Speech Separation},
author = {Jisi Zhang and Catalin Zorila and Rama Doddipatla and Jon Barker},
journal= {arXiv preprint arXiv:2106.07843},
year = {2021}
}
Comments
Accepted to Interspeech 2021