Automatic Speech Recognition
Transformers
PyTorch
TensorFlow
JAX
Safetensors
whisper
audio
hf-asr-leaderboard
Eval Results (legacy)
Instructions to use openai/whisper-medium with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use openai/whisper-medium with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="openai/whisper-medium")# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("openai/whisper-medium") model = AutoModelForSpeechSeq2Seq.from_pretrained("openai/whisper-medium") - Notebooks
- Google Colab
- Kaggle
Add `return_timestamps` attribute to generation config
#29
by sanchit-gandhi - opened
For consistency with the other Whisper models, we should set the return_timestamps attribute to False in the generation config, e.g.: for Whisper Small: https://huggingface.co/openai/whisper-small/blob/e34e8ae444c29815eca53e11383ea13b2e362eb0/generation_config.json#L167
cc @patrickvonplaten if you could merge that would be grand! Thanks!
patrickvonplaten changed pull request status to merged