Update app.py

app.py

@@ -274,27 +274,9 @@ def initialize_agent_with_prompt(prompt_template):
     return agent
 
 
-# def generate_answer(message, choice):
-#     logging.debug(f"generate_answer called with prompt_choice: {choice}")
-    
-#     if choice == "Details":
-#         agent = initialize_agent_with_prompt(QA_CHAIN_PROMPT_1)
-#     elif choice == "Conversational":
-#         agent = initialize_agent_with_prompt(QA_CHAIN_PROMPT_2)
-#     else:
-#         logging.error(f"Invalid prompt_choice: {choice}. Defaulting to 'Conversational'")
-#         agent = initialize_agent_with_prompt(QA_CHAIN_PROMPT_2)
-#     response = agent(message)
-
-#     addresses = extract_addresses(response['output'])
-#     return response['output'], addresses
-
 def generate_answer(message, choice):
     logging.debug(f"generate_answer called with prompt_choice: {choice}")
     
-    # Reset conversational memory
-    conversational_memory.clear()
-    
     if choice == "Details":
         agent = initialize_agent_with_prompt(QA_CHAIN_PROMPT_1)
     elif choice == "Conversational":
@@ -309,6 +291,8 @@ def generate_answer(message, choice):
 
 
 
+
+
 def bot(history, choice, tts_choice, state):
     if not history:
         return history
@@ -495,7 +479,21 @@ base_audio_drive = "/data/audio"
 #     return stream, full_text, result
 
 
+# def transcribe_function(stream, new_chunk):
+#     sr, y = new_chunk[0], new_chunk[1]
+#     y = y.astype(np.float32) / np.max(np.abs(y))
+#     if stream is not None:
+#         stream = np.concatenate([stream, y])
+#     else:
+#         stream = y
+#     result = pipe_asr({"array": stream, "sampling_rate": sr}, return_timestamps=False)
+#     full_text = result.get("text", "")
+#     return stream, full_text  # Return the transcribed text
+
 def transcribe_function(stream, new_chunk):
+    # Reset conversational memory
+    conversational_memory.clear()
+    
     sr, y = new_chunk[0], new_chunk[1]
     y = y.astype(np.float32) / np.max(np.abs(y))
     if stream is not None:
@@ -504,7 +502,7 @@ def transcribe_function(stream, new_chunk):
         stream = y
     result = pipe_asr({"array": stream, "sampling_rate": sr}, return_timestamps=False)
     full_text = result.get("text", "")
-    return stream, full_text  # Return the transcribed text
+    return stream, full_text
 
 def update_map_with_response(history):
     if not history:
@@ -711,9 +709,7 @@ with gr.Blocks(theme='Pijush2023/scikit-learn-pijush') as demo:
             clear_button.click(fn=clear_textbox, inputs=None, outputs=chat_input)
            
             audio_input = gr.Audio(sources=["microphone"], streaming=True, type='numpy')
-            audio_input.stream(transcribe_function, inputs=[state, audio_input], outputs=[state, chat_input], api_name="voice_query_to_text")
-
-            
+            audio_input.stream(transcribe_function, inputs=[state, audio_input], outputs=[state, chat_input], api_name="voice_query_to_text")