Wav2Vec2 Base 960h

JSON →
meta audio
audio

A base-sized Wav2Vec2 model pretrained on 960 hours of LibriSpeech for English speech recognition.

streaming
releasedJun 2021