Chen, et al.. Speech Representation Learning Through Self-supervised Pretraining and Multi-task Finetuning. 18 Oct. 2021.