Fetching the paper…

vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations · Around