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
| 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 | |
| ``` |