W2V-BERT 2.0

JSON →
meta audio
audio

A self-supervised speech representation model from Facebook combining wav2vec 2.0 and BERT for audio understanding.

open-weights
releasedJun 2024