Models / meta / Wav2Vec2 Base 960h

Wav2Vec2 Base 960h

JSON →
metaaudio

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

audiostreaming
homepage ↗
Specs
Context & limits
—
context window
—
max output
Lifecycle
Dates
released2021-06-01
Resources
homepagehttps://huggingface.co/facebook/wav2vec2-base-960h ↗