Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
whisper
whisper-event
Generated from Trainer
hf-asr-leaderboard
Eval Results (legacy)
Instructions to use arbml/whisper-largev2-ar with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use arbml/whisper-largev2-ar with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="arbml/whisper-largev2-ar")# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("arbml/whisper-largev2-ar") model = AutoModelForSpeechSeq2Seq.from_pretrained("arbml/whisper-largev2-ar") - Notebooks
- Google Colab
- Kaggle
How come the model is only 3GB?
#1
by nouamanetazi - opened
If this model is fine-tuned from https://huggingface.co/openai/whisper-large-v2 , and the latter is 6GB
Why is the pytorch_model.bin in this repo only 3GB? π€
Great job btw!
cc @Zaid
We used deepspeed for training this model. The reduction in size might ve due to the optmiziation.