Update app.py

app.py

@@ -183,59 +183,54 @@ with gr.Blocks(title="Vevo Voice Conversion") as demo:
             label="Inference Mode",
             interactive=True
         )
-    
-    # Audio inputs
-    content_audio = gr.Audio(
-        label="Source Audio",
-        type="filepath",
-        visible=True,
-        interactive=True
-    )
-    
-    ref_style_audio = gr.Audio(
-        label="Reference Style Audio",
-        type="filepath",
-        visible=True,
-        interactive=True
-    )
-    
-    ref_timbre_audio = gr.Audio(
-        label="Reference Timbre Audio",
-        type="filepath",
-        visible=True,
-        interactive=True
-    )
-    
-    # Text inputs
-    src_text = gr.Textbox(
-        label="Source Text",
-        placeholder="Enter text for TTS mode",
-        visible=False,
-        interactive=True
-    )
-    
-    ref_text = gr.Textbox(
-        label="Reference Style Text",
-        placeholder="Optional: Enter reference text",
-        visible=False,
-        interactive=True
-    )
-    
-    src_language = gr.Dropdown(
-        choices=["en", "zh"],
-        value="en",
-        label="Source Language",
-        visible=False,
-        interactive=True
-    )
-    
-    ref_language = gr.Dropdown(
-        choices=["en", "zh"],
-        value="en",
-        label="Reference Language",
-        visible=False,
-        interactive=True
-    )
+
+    with gr.Tabs():
+        with gr.TabItem("Audio Inputs"):
+            content_audio = gr.Audio(
+                label="Source Audio",
+                type="filepath",
+                interactive=True
+            )
+            
+            ref_style_audio = gr.Audio(
+                label="Reference Style Audio",
+                type="filepath",
+                interactive=True
+            )
+            
+            ref_timbre_audio = gr.Audio(
+                label="Reference Timbre Audio",
+                type="filepath",
+                interactive=True
+            )
+        
+        with gr.TabItem("Text Inputs (TTS Mode)"):
+            src_text = gr.Textbox(
+                label="Source Text",
+                placeholder="Enter text for TTS mode",
+                interactive=True
+            )
+            
+            ref_text = gr.Textbox(
+                label="Reference Style Text (Optional)",
+                placeholder="Enter reference text",
+                interactive=True
+            )
+            
+            with gr.Row():
+                src_language = gr.Dropdown(
+                    choices=["en", "zh"],
+                    value="en",
+                    label="Source Language",
+                    interactive=True
+                )
+                
+                ref_language = gr.Dropdown(
+                    choices=["en", "zh"],
+                    value="en",
+                    label="Reference Language",
+                    interactive=True
+                )
     
     with gr.Row():
         steps = gr.Slider(
@@ -247,43 +242,22 @@ with gr.Blocks(title="Vevo Voice Conversion") as demo:
         )
     
     with gr.Row():
-        submit_btn = gr.Button("Generate")
+        with gr.Column():
+            submit_btn = gr.Button("Generate")
+            error_box = gr.Textbox(label="Status", interactive=False)
         output_audio = gr.Audio(label="Generated Audio")
     
-    # Handle visibility of components based on mode
-    def update_interface(mode):
-        is_tts = mode == "tts"
-        is_timbre = mode == "timbre"
-        
-        return [
-            # Audio components visibility
-            not is_tts,  # content_audio
-            not is_timbre and not is_tts,  # ref_style_audio
-            True,  # ref_timbre_audio (always visible)
-            # Text components visibility
-            is_tts,  # src_text
-            is_tts,  # ref_text
-            is_tts,  # src_language
-            is_tts   # ref_language
-        ]
-    
-    mode.change(
-        fn=update_interface,
-        inputs=[mode],
-        outputs=[
-            content_audio,
-            ref_style_audio,
-            ref_timbre_audio,
-            src_text,
-            ref_text,
-            src_language,
-            ref_language
-        ]
-    )
+    def process_with_error_handling(*args):
+        try:
+            result = process_audio(*args)
+            error_box.update(value="Success!")
+            return [result, "Success!"]
+        except Exception as e:
+            error_msg = str(e)
+            return [None, error_msg]
     
-    # Handle generation
     submit_btn.click(
-        fn=process_audio,
+        fn=process_with_error_handling,
         inputs=[
             mode,
             content_audio,
@@ -295,8 +269,33 @@ with gr.Blocks(title="Vevo Voice Conversion") as demo:
             ref_language,
             steps
         ],
-        outputs=output_audio
+        outputs=[output_audio, error_box]
     )
 
+    # Example usage text
+    gr.Markdown("""
+    ## Quick Start Guide
+    
+    1. Select your mode:
+       - **Voice**: Full voice conversion (style + timbre)
+       - **Timbre**: Only voice timbre conversion
+       - **TTS**: Text-to-speech with voice cloning
+    
+    2. For Voice/Timbre modes:
+       - Upload source audio (what you want to convert)
+       - Upload reference audio(s)
+    
+    3. For TTS mode:
+       - Enter your text
+       - Select language
+       - Upload reference audio(s)
+    
+    4. Adjust steps slider:
+       - Higher values = better quality but slower
+       - Lower values = faster but lower quality
+    
+    5. Click Generate and wait for processing
+    """)
+
 if __name__ == "__main__":
     demo.queue().launch()