facebook/wav2vec2-large-960h-lv60-self

Automatic Speech Recognition • Updated May 23, 2022 • 96.6k • 130

Note The Wav2Vec 2.0 "large" model pre-trained on 53k hours of un-labelled audio data from the LibriSpeech and LibriVox (LV) corpora, and fine-tuned on 960 hours of LibriSpeech ASR data. This is the most performant Wav2Vec 2.0 checkpoint from the initial release, obtaining 1.9/3.9% WER on the LibriSpeech test clean/other subsets respectively.

facebook/wav2vec2-large-960h

Automatic Speech Recognition • Updated Apr 5, 2022 • 72.5k • 26

Note The Wav2Vec 2.0 "large" model pre-trained and fine-tuned on 960 hours of LibriSpeech ASR data.

facebook/wav2vec2-base-960h

Automatic Speech Recognition • Updated Nov 14, 2022 • 1.24M • 306

Note The Wav2Vec 2.0 "base" model pre-trained and fine-tuned on 960 hours of LibriSpeech ASR data.

facebook/wav2vec2-base-100h

Automatic Speech Recognition • Updated May 27, 2022 • 2.04k • 6

Note The Wav2Vec 2.0 "base" model pre-trained on 960 hours of un-labelled LibriSpeech ASR data, and fine-tuned on 100 hours of labelled LibriSpeech ASR data.

facebook/wav2vec2-large-lv60

Updated Dec 28, 2021 • 15.4k • 6

Note The Wav2Vec 2.0 "large" model pre-trained on 53k hours of un-labelled data from the LibriSpeech and LibriVox (LV) corpora.

facebook/wav2vec2-large

Updated Aug 26, 2022 • 3.64k • 3

Note The Wav2Vec 2.0 "large" model pre-trained on 960 hours of un-labelled LibriSpeech ASR data.

facebook/wav2vec2-base

Updated Dec 28, 2021 • 1.01M • 77

Note The Wav2Vec 2.0 "base" model pre-trained on 960 hours of un-labelled LibriSpeech ASR data.

Note The wav2vec 2.0 paper, accepted to NeurIPS 2020.