Update app.py

app.py

@@ -43,9 +43,11 @@ def inference(audio):
     options = whisper.DecodingOptions(fp16 = False)
     # transcribe speech to text
     result = whisper.decode(model, mel, options)
-
+    
+    # Added prompt two 2 lines below
+    prompt = "The following is a transcript of someone talking, please predict what they will say next. \n"
     ### code 
-    input_ids = tokenizer(result.text, return_tensors="pt").input_ids
+    input_ids = tokenizer(prompt, result.text, return_tensors="pt").input_ids
     print("inputs ", input_ids)
     # length penalty for gpt2.generate??? 
     generated_outputs = gpt2.generate(input_ids, do_sample=True, num_return_sequences=3, output_scores=True)