Instructions to use utter-project/mHuBERT-147 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use utter-project/mHuBERT-147 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="utter-project/mHuBERT-147")# Load model directly from transformers import AutoProcessor, AutoModel processor = AutoProcessor.from_pretrained("utter-project/mHuBERT-147") model = AutoModel.from_pretrained("utter-project/mHuBERT-147") - Inference
- Notebooks
- Google Colab
- Kaggle
Error while loading the checkpoint with fairseq
Hi!
I am trying to make a continuous-pretrain from this checkpoint, but I am struggling to load the .pt checkpoint with your fork of Fairseq. I get this error:
AssertionError: Could not infer task type from {'_name': 'mhubert_pretraining' ...}
It seems that there is not a task named 'mhubert_pretrain'. Do you know how can I solve this?
Thank you in advance!
Hi Ansu,
Thanks for the interest in the model. This is curious, because the task should indeed be there: https://github.com/utter-project/fairseq/blob/2959684ecc6e28e34a075e2f1a08bba100e11743/fairseq/tasks/hubert_pretraining.py#L293
Did you make sure to pip install -e . inside the fairseq fork repository before trying to run the training?
Hi,
Sorry for the late response. Yeah, that was the problem, I had installed the original fairseq as I couldn't install your fork, but I got it working.
Thanks!