Added clean main tab with voice-first interaction

app.py

@@ -465,6 +465,7 @@ def process_turn(
             None,
             None,
             None,
+            None,
         )
 
     patient_text = (text_input or "").strip()
@@ -489,6 +490,7 @@ def process_turn(
             None,
             None,
             None,
+            None,
         )
 
     # Add patient's message
@@ -561,6 +563,7 @@ def process_turn(
         None,
         None,
         tts_path,
+        tts_path,
     )
 
 
@@ -586,15 +589,16 @@ def create_demo():
         )
 
         with gr.Tabs():
-            with gr.TabItem("Chat"):
+            with gr.TabItem("Main"):
                 with gr.Column():
-                    chatbot = gr.Chatbot(height=440, type="tuples", label="Conversation")
+                    chatbot = gr.Chatbot(height=480, type="tuples", label="Conversation")
                     with gr.Row():
-                        audio = gr.Audio(sources=["microphone"], type="filepath", label="Speak your response (or type below)")
-                        text = gr.Textbox(lines=2, placeholder="Type your message (optional if you speak)")
+                        audio_main = gr.Audio(sources=["microphone"], type="filepath", label="Speak")
+                    text_main = gr.Textbox(lines=1, placeholder="Type a message and press Enter")
                     with gr.Row():
-                        send_btn = gr.Button("Send")
-                        reset_btn = gr.Button("Reset")
+                        reset_btn_main = gr.Button("Reset")
+                    tts_audio_main = gr.Audio(label="Clinician voice", interactive=False, autoplay=True, show_label=False)
+        
             with gr.TabItem("Advanced"):
                 with gr.Column():
                     score_json = gr.JSON(label="PHQ-9 Assessment (live)")
@@ -619,15 +623,22 @@ def create_demo():
         demo.load(_on_load_init, inputs=None, outputs=[chatbot, scores_state, meta_state, finished_state, turns_state])
 
         # Wire interactions
-        send_btn.click(
+        audio_main.change(
             fn=process_turn,
-            inputs=[audio, text, chatbot, threshold, tts_enable, finished_state, turns_state, scores_state, meta_state],
-            outputs=[chatbot, score_json, severity_label, finished_state, turns_state, audio, text, tts_audio],
+            inputs=[audio_main, text_main, chatbot, threshold, tts_enable, finished_state, turns_state, scores_state, meta_state],
+            outputs=[chatbot, score_json, severity_label, finished_state, turns_state, audio_main, text_main, tts_audio, tts_audio_main],
             queue=True,
             api_name="message",
         )
 
-        reset_btn.click(fn=reset_app, inputs=None, outputs=[chatbot, scores_state, meta_state, finished_state, turns_state])
+        text_main.submit(
+            fn=process_turn,
+            inputs=[audio_main, text_main, chatbot, threshold, tts_enable, finished_state, turns_state, scores_state, meta_state],
+            outputs=[chatbot, score_json, severity_label, finished_state, turns_state, audio_main, text_main, tts_audio, tts_audio_main],
+            queue=True,
+        )
+
+        reset_btn_main.click(fn=reset_app, inputs=None, outputs=[chatbot, scores_state, meta_state, finished_state, turns_state])
 
         # Model switch handlers
         def _on_apply_model(mid: str):