Instructions to use TechInterMezzo/whisper-encoder-medium with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use TechInterMezzo/whisper-encoder-medium with Transformers:
# Load model directly from transformers import AutoProcessor, WhisperEncoder processor = AutoProcessor.from_pretrained("TechInterMezzo/whisper-encoder-medium") model = WhisperEncoder.from_pretrained("TechInterMezzo/whisper-encoder-medium", device_map="auto") - Notebooks
- Google Colab
- Kaggle
File size: 552 Bytes
5f7e788 37f94c5 1609f1f 37f94c5 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | ---
license: mit
---
```python
from transformers import WhisperFeatureExtractor
from transformers.models.whisper.modeling_whisper import WhisperEncoder
feature_extractor = WhisperFeatureExtractor.from_pretrained("techintermezzo/whisper-encoder-medium")
model = WhisperEncoder.from_pretrained("techintermezzo/whisper-encoder-medium").half()
model.eval()
with torch.inference_mode():
input_features = feature_extractor(inputs, sampling_rate=16000, return_tensors="pt").input_features
last_hidden_state = model(input_features).last_hidden_state
``` |