Instructions to use nyralabs/CrisperWhisper with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use nyralabs/CrisperWhisper with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="nyralabs/CrisperWhisper")# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("nyralabs/CrisperWhisper") model = AutoModelForSpeechSeq2Seq.from_pretrained("nyralabs/CrisperWhisper") - Notebooks
- Google Colab
- Kaggle
Smaller versions?
#7
by hammeiam - opened
Hi again,
I wonder if it's on your roadmap to train different sizes of the model? Would love to have some smaller options to choose from like whisper-small, whisper-medium etc.
Hey, unfortunately as of yet this is not on the roadmap. We plan to release a upgraded version of the model trained on much more verbatim data, especially for german and introduce a turbo-version of the model. Smaller variants are not on the roadmap yet.