Wav2Vec2 model pre-trained using the Spanish portion of the Common Voice dataset. The model is trained with Flax and using TPUs sponsored by Google since this is part of the
Flax/Jax Community Week organised by HuggingFace.
The model used for training is
Wav2Vec2 by FacebookAI. It was introduced in the paper
"wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations" by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, and Michael Auli (
https://arxiv.org/abs/2006.11477).
This model is available in the 🤗
Model Hub.
Spanish portion of
Common Voice. Common Voice is an open source, multi-language dataset of voices part of Mozilla's initiative to help teach machines how real people speak.
The dataset is also available in the 🤗
Datasets library.
