Update app.py

app.py

@@ -47,7 +47,8 @@ def inference(audio):
     # Added prompt below
     input_prompt = "The following is a transcript of someone talking, please predict what they will say next. \n"
     ### code 
-    input_ids = tokenizer(input_prompt, result.text, return_tensors="pt").input_ids
+    input_total = input_prompt + result.text
+    input_ids = tokenizer(input_total, return_tensors="pt").input_ids
     print("inputs ", input_ids)
 
     # prompt length