Update app.py

app.py

@@ -45,7 +45,7 @@ def generate_llm_response(text, model_id="ccibeekeoc42/Llama3.1-8b-base-SFT-2024
             ],
             top_p=0.3,
             temperature=1,
-            max_tokens=80,
+            max_tokens=150,
             stream=True,
             seed=None,
             stop=None,
@@ -226,7 +226,7 @@ def speech_to_speech_translation(audio):
     synthesised_speech = (synthesised_speech.numpy() * max_range).astype(np.int16)
     print("Speech Synthesis Completed~~~~~~~~~~~~~~~~~~~")
 
-    return transcribed_text, (16000, synthesised_speech), llm_response
+    return transcribed_text, llm_response, (16000, synthesised_speech)
 
 
 # Gradio Demo