Try to save tokenizer

app.py

@@ -12,6 +12,13 @@ BATCH_SIZE = 8
 FILE_LIMIT_MB = 1000
 YT_LENGTH_LIMIT_S = 3600  # limit to 1 hour YouTube files
 
+from transformers import AutoTokenizer
+
+tokenizer = AutoTokenizer.from_pretrained("openai/whisper-large-v3")
+assert tokenizer.is_fast
+tokenizer.save_pretrained("...")
+
+
 device = 0 if torch.cuda.is_available() else "cpu"
 
 pipe = pipeline(