How to use marczenko/whisper_tokenizer with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("marczenko/whisper_tokenizer", dtype="auto")
