Hugging Face's logo

facebook
/
wav2vec2-large-100k-voxpopuli

Automatic Speech Recognition
Transformers
PyTorch
JAX
multilingual
wav2vec2
pretraining
audio
voxpopuli
Inference Endpoints
Model card Files Community
2
wav2vec2-large-100k-voxpopuli / README.md
patrickvonplaten's picture
patrickvonplaten
Update README.md
ad2f1b5
preview code
|
raw
history blame contribute delete
1.42 kB
metadata
language: multilingual
tags:
  - audio
  - automatic-speech-recognition
  - voxpopuli
license: cc-by-nc-4.0

Wav2Vec2-Large-VoxPopuli

Facebook's Wav2Vec2 large model pretrained on the 100k unlabeled subset of VoxPopuli corpus.

Note: This model does not have a tokenizer as it was pretrained on audio alone. In order to use this model speech recognition, a tokenizer should be created and the model should be fine-tuned on labeled text data. Check out this blog for more in-detail explanation of how to fine-tune the model.

Paper: VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation

Authors: Changhan Wang, Morgane Riviere, Ann Lee, Anne Wu, Chaitanya Talnikar, Daniel Haziza, Mary Williamson, Juan Pino, Emmanuel Dupoux from Facebook AI

See the official website for more information, here

Fine-Tuning

Please refer to this blog on how to fine-tune this model on a specific language. Note that you should replace "facebook/wav2vec2-large-xlsr-53" with this checkpoint for fine-tuning.