HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units

Open in new window