Automatic speech quality assessment is an important, transversal task whose progress is hampered by the scarcity of human annotations, poor generalization to unseen recording conditions, and a lack of flexibility of existing approaches. In this work, we tackle these problems with a semi-supervised learning approach, combining available annotations with programmatically generated data, and using 3 different optimization criteria together with 5 complementary auxiliary tasks. Our results show that such a semi-supervised approach can cut the error of existing methods by more than 36%, while providing additional benefits in terms of reusable features or auxiliary outputs. Improvement is further corroborated with an out-of-sample test showing promising generalization capabilities.
@article{arxiv.2010.00368,
title = {SESQA: semi-supervised learning for speech quality assessment},
author = {Joan Serrà and Jordi Pons and Santiago Pascual},
journal= {arXiv preprint arXiv:2010.00368},
year = {2021}
}
Comments
Long version (with appendix) of the paper with the same title accepted for ICASSP2021