Try fix UI bug

app.py

@@ -2,13 +2,11 @@ import gradio as gr
 import outetts
 import json
 import tempfile
-import os
-from typing import Optional, Dict, Any, List, Tuple
 
 # Available OuteTTS models based on the documentation
 MODELS = {v.value: v for _, v in outetts.Models.__members__.items()}
 
-def initialize_interface(model_name: str) -> Any:
+def initialize_interface(model_name: str):
     """Initialize the OuteTTS interface with selected model."""
     model = MODELS[model_name]
     
@@ -23,146 +21,86 @@ def initialize_interface(model_name: str) -> Any:
     interface = outetts.Interface(config=config)
     return interface
 
-def create_speaker_from_audio(audio_file, interface_state) -> Tuple[str, Any]:
-    """Create a speaker profile from uploaded audio file."""
-    if interface_state is None:
-        raise gr.Error("Please select a model first")
-    
+def create_speaker_and_generate(model_name, audio_file, test_text="", temperature=0.4):
+    """Create speaker from audio and optionally generate test audio."""
     if audio_file is None:
         raise gr.Error("Please upload an audio file")
     
+    # Initialize model
+    interface = initialize_interface(model_name)
+    
     # Create speaker profile from audio
-    speaker = interface_state.create_speaker(audio_file)
+    speaker = interface.create_speaker(audio_file)
     
     # Convert speaker dict to formatted JSON
     speaker_json = json.dumps(speaker, indent=2, ensure_ascii=False)
     
-    return speaker_json, speaker
-
-
-
-def generate_sample_audio(text: str, temperature: float, interface_state, speaker_state) -> Optional[str]:
-    """Generate sample audio using the current speaker."""
-    if interface_state is None:
-        raise gr.Error("Please select a model first")
-    
-    if speaker_state is None:
-        raise gr.Error("Please create a speaker first")
-    
-    if not text.strip():
-        raise gr.Error("Please enter text to generate")
-    
-    # Generate audio
-    output = interface_state.generate(
-        config=outetts.GenerationConfig(
-            text=text,
-            speaker=speaker_state,
-            sampler_config=outetts.SamplerConfig(
-                temperature=temperature
+    # Generate test audio if text is provided
+    generated_audio = None
+    if test_text and test_text.strip():
+        output = interface.generate(
+            config=outetts.GenerationConfig(
+                text=test_text,
+                speaker=speaker,
+                sampler_config=outetts.SamplerConfig(
+                    temperature=temperature
+                )
             )
         )
-    )
+        
+        # Save to temporary file
+        with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f:
+            output.save(f.name)
+            generated_audio = f.name
     
-    # Save to temporary file
-    with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f:
-        output.save(f.name)
-        return f.name
+    return speaker_json, generated_audio
 
 # Create the Gradio interface
-with gr.Blocks(title="OuteTTS Speaker Creator", theme=gr.themes.Soft()) as demo:
-    # Add spacing to avoid Hugging Face header overlap
-    gr.HTML("<div style='height: 200px;'></div>")
-    
-    with gr.Row():
-        gr.Markdown("""
-        # 🎙️ OuteTTS Speaker Creator
-        Create and manage speaker profiles for OuteTTS text-to-speech synthesis
-        """)
-    
-    gr.HTML("<div style='height: 10px;'></div>")
-    
-    # State management for multi-user support
-    interface_state = gr.State(value=None)
-    speaker_state = gr.State(value=None)
-    
-    with gr.Row():
-        with gr.Column(scale=1):
-            gr.Markdown("## 🔧 Model Selection")
-            
-            model_dropdown = gr.Dropdown(
-                choices=list(MODELS.keys()),
-                value=list(MODELS.keys())[-1],  # Default to last model
-                label="Select OuteTTS Model",
-                info="Choose the model variant to use"
-            )
-            
-        with gr.Column(scale=1):
-            gr.Markdown("## 🎵 Speaker Creation")
-            
-            audio_upload = gr.Audio(
-                label="Upload Reference Audio",
-                type="filepath",
-                sources=["upload", "microphone"]
-            )
-            create_button = gr.Button("Create Speaker Profile", variant="primary")
-    
-    with gr.Row():
-        with gr.Column():
-            gr.Markdown("## 📋 Speaker Profile (JSON)")
-            speaker_json = gr.Textbox(
-                label="Speaker JSON",
-                lines=15,
-                max_lines=20,
-                interactive=False,
-                show_copy_button=True
-            )
-    
-    with gr.Row():
-        with gr.Column():
-            gr.Markdown("## 🎧 Test Audio Generation")
-            test_text = gr.Textbox(
-                label="Test Text",
-                placeholder="Enter text to generate speech...",
-                lines=3,
-                value="Hello, this is a test of the OuteTTS speaker profile."
-            )
-            
-            temperature_slider = gr.Slider(
-                minimum=0.1,
-                maximum=1.0,
-                step=0.1,
-                value=0.4,
-                label="Temperature",
-                info="Controls randomness in generation"
-            )
-            
-            generate_button = gr.Button("Generate Sample Audio", variant="primary")
-            sample_audio = gr.Audio(label="Generated Audio", interactive=False)
-    
-    # Event handlers
-    # Initialize default model on page load
-    demo.load(
-        fn=initialize_interface,
-        inputs=[model_dropdown],
-        outputs=[interface_state]
-    )
-    
-    model_dropdown.change(
-        fn=initialize_interface,
-        inputs=[model_dropdown],
-        outputs=[interface_state]
-    )
-    
-    create_button.click(
-        fn=create_speaker_from_audio,
-        inputs=[audio_upload, interface_state],
-        outputs=[speaker_json, speaker_state]
-    )
-    
-    generate_button.click(
-        fn=generate_sample_audio,
-        inputs=[test_text, temperature_slider, interface_state, speaker_state],
-        outputs=[sample_audio]
-    )
+demo = gr.Interface(
+    fn=create_speaker_and_generate,
+    inputs=[
+        gr.Dropdown(
+            choices=list(MODELS.keys()),
+            value=list(MODELS.keys())[-1],
+            label="Select OuteTTS Model",
+            info="Choose the model variant to use"
+        ),
+        gr.Audio(
+            label="Upload Reference Audio",
+            type="filepath",
+            sources=["upload", "microphone"]
+        ),
+        gr.Textbox(
+            label="Test Text (Optional)",
+            placeholder="Enter text to generate speech (leave empty to only create speaker profile)...",
+            lines=3,
+            value="Hello, this is a test of the OuteTTS speaker profile."
+        ),
+        gr.Slider(
+            minimum=0.1,
+            maximum=1.0,
+            step=0.1,
+            value=0.4,
+            label="Temperature",
+            info="Controls randomness in generation"
+        )
+    ],
+    outputs=[
+        gr.Textbox(
+            label="Speaker Profile (JSON)",
+            lines=15,
+            max_lines=20,
+            show_copy_button=True
+        ),
+        gr.Audio(
+            label="Generated Test Audio (if text provided)",
+            type="filepath"
+        )
+    ],
+    title="🎙️ OuteTTS Speaker Creator",
+    description="Create and manage speaker profiles for OuteTTS text-to-speech synthesis. Upload audio to create a speaker profile, and optionally provide test text to generate sample audio.",
+    theme=gr.themes.Soft(),
+    examples=None
+)
 
 demo.launch()