Instructions to use facebook/w2v-bert-2.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/w2v-bert-2.0 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="facebook/w2v-bert-2.0")# Load model directly from transformers import AutoProcessor, AutoModel processor = AutoProcessor.from_pretrained("facebook/w2v-bert-2.0") model = AutoModel.from_pretrained("facebook/w2v-bert-2.0") - Notebooks
- Google Colab
- Kaggle
Is the default padding value incorrect?
#31
by Devilstro - opened
We've been using different audio encoder models, and had troubles with the Wav2Vec2 BERT model specifically - then we realised the model defaults to padding with 1.0 as opposed to 0.0 (like all other encoders).
This seems unintentional / incorrect when working with spectograms.