Update app.py

app.py

@@ -232,7 +232,7 @@ def synthesize(text, reference_audio, language="en", speed=1.0, progress=gr.Prog
         if "list index out of range" in error or "index out of range" in error:
             raise gr.Error(
                 "Voice extraction failed.\n\n"
-                "✅ SOLUTION: Use microphone recording:\n"
+                "SOLUTION: Use microphone recording:\n"
                 "1. Click the microphone icon\n"
                 "2. Record 6-10 seconds clearly\n"
                 "3. Say: 'Hello, this is my test voice'\n"
@@ -335,16 +335,16 @@ def clear_cache():
     _tts = None
     if torch.cuda.is_available():
         torch.cuda.empty_cache()
-    return "✅ Cache cleared"
+    return "Cache cleared"
 
 def get_status():
     if _tts is not None:
-        return "✅ Model ready (GPU)" if torch.cuda.is_available() else "✅ Model ready (CPU)"
+        return "Model ready (GPU)" if torch.cuda.is_available() else "Model ready (CPU)"
     elif _model_loading:
-        return "⏳ Loading model..."
+        return "Loading model..."
     elif _model_error:
-        return f"❌ Error"
-    return "🔄 Ready - first use loads model"
+        return "Error"
+    return "Ready - model loads on first use"
 
 def estimate_duration(text, speed=1.0):
     if not text:
@@ -356,73 +356,57 @@ def estimate_duration(text, speed=1.0):
     return f"{mins}m {secs}s" if mins > 0 else f"{secs}s"
 
 # ============================================================
-# UI - COMPLETE WITH ALL FEATURES
+# UI - NO TRIPLE QUOTES
 # ============================================================
 
-with gr.Blocks(title="VoxForge TTS", theme=gr.themes.Soft(), css=".gradio-container {max-width: 1200px !important; margin: auto !important;}") as demo:
+with gr.Blocks(title="VoxForge TTS", theme=gr.themes.Soft()) as demo:
     
     gr.Markdown("# 🎙️ VoxForge TTS - Professional Voice Cloning")
-    gr.Markdown("**Upload any voice or use microphone** to record 6-10 seconds. First use downloads model (2-5 min).")
+    gr.Markdown("Upload any voice or use microphone to record 6-10 seconds. First use downloads model (2-5 min).")
     
     with gr.Tabs():
         
-        # ========================================================
-        # TAB 1: STANDARD SYNTHESIS
-        # ========================================================
-        with gr.Tab("🎤 Standard Synthesis"):
+        # TAB 1: Standard Synthesis
+        with gr.Tab("Standard Synthesis"):
             with gr.Row():
                 with gr.Column(scale=1):
-                    text_input = gr.Textbox(
-                        label="Text to Synthesize",
-                        lines=6,
-                        max_length=2000,
-                        placeholder="Enter text up to 2000 characters...\n\nExample: Hello, this is my cloned voice speaking naturally."
-                    )
+                    text_input = gr.Textbox(label="Text to Synthesize", lines=6, max_length=2000, placeholder="Enter text up to 2000 characters... Example: Hello, this is my cloned voice.")
                     
                     with gr.Row():
                         char_count = gr.Label("0/2000")
                         duration_est = gr.Label("Est. 0s")
                     
-                    ref_audio = gr.Audio(
-                        label="Reference Voice (6-10 seconds - Click microphone to record!)",
-                        type="filepath",
-                        sources=["upload", "microphone"]
-                    )
+                    ref_audio = gr.Audio(label="Reference Voice (6-10 seconds - Click microphone to record!)", type="filepath", sources=["upload", "microphone"])
                     
                     with gr.Row():
                         language = gr.Dropdown(choices=SUPPORTED_LANGUAGES, value="en", label="Language")
                         speed = gr.Slider(0.5, 2.0, value=1.0, step=0.05, label="Speed")
                     
                     with gr.Row():
-                        gen_btn = gr.Button("🎵 Generate Speech", variant="primary", size="lg")
-                        clear_btn = gr.Button("🗑️ Clear", variant="secondary", size="lg")
+                        gen_btn = gr.Button("Generate Speech", variant="primary", size="lg")
+                        clear_btn = gr.Button("Clear", variant="secondary", size="lg")
                     
-                    with gr.Accordion("⚙️ Advanced", open=False):
+                    with gr.Accordion("Advanced Options", open=False):
                         status_text = gr.Label(get_status())
                         clear_cache_btn = gr.Button("Clear Model Cache", size="sm")
-                        clear_output = gr.Textbox(visible=False)
                 
                 with gr.Column(scale=1):
-                    audio_output = gr.Audio(label="🔊 Generated Speech", type="filepath")
+                    audio_output = gr.Audio(label="Generated Speech", type="filepath")
                     
-                    gr.Markdown("""
-                    ### ⭐ Tips for Best Results
-                    
-                    1. **Use microphone** (click the mic icon) - records perfectly every time!
-                    2. **Speak clearly** for 6-10 seconds
-                    3. **No background noise** - one speaker only
-                    4. **Match language** to your voice
-                    
-                    ### 🚀 Speed Tips
-                    - First synthesis: 2-5 min (downloads 4GB model)
-                    - After that: 10-30 seconds
-                    - Enable GPU in Space settings for faster results
-                    
-                    ### 🌐 Supported Languages
-                    English, Spanish, French, German, Italian, Portuguese, Polish, Turkish, Russian, Dutch, Czech, Arabic, Chinese, Japanese
-                    """)
+                    gr.Markdown("### Tips for Best Results")
+                    gr.Markdown("1. Use microphone (click the mic icon) - records perfectly every time!")
+                    gr.Markdown("2. Speak clearly for 6-10 seconds")
+                    gr.Markdown("3. No background noise - one speaker only")
+                    gr.Markdown("4. Match language to your voice")
+                    gr.Markdown("")
+                    gr.Markdown("### Performance Notes")
+                    gr.Markdown("- First synthesis: 2-5 min (downloads 4GB model)")
+                    gr.Markdown("- After that: 10-30 seconds")
+                    gr.Markdown("- Enable GPU in Space settings for faster results")
+                    gr.Markdown("")
+                    gr.Markdown("### Supported Languages")
+                    gr.Markdown("English, Spanish, French, German, Italian, Portuguese, Polish, Turkish, Russian, Dutch, Czech, Arabic, Chinese, Japanese")
             
-            # Event handlers
             def update_char(t):
                 return f"{len(t)}/2000" if t else "0/2000"
             
@@ -435,77 +419,75 @@ with gr.Blocks(title="VoxForge TTS", theme=gr.themes.Soft(), css=".gradio-contai
             
             gen_btn.click(synthesize, [text_input, ref_audio, language, speed], [audio_output])
             clear_btn.click(lambda: ("", None, "en", 1.0), None, [text_input, ref_audio, language, speed])
-            clear_cache_btn.click(clear_cache, None, [clear_output]).then(lambda: get_status(), None, [status_text])
+            clear_cache_btn.click(clear_cache, None, [status_text]).then(lambda: get_status(), None, [status_text])
         
-        # ========================================================
-        # TAB 2: BATCH SYNTHESIS
-        # ========================================================
-        with gr.Tab("📚 Batch (Long Text)"):
-            gr.Markdown("### 🔄 For texts over 2000 characters")
+        # TAB 2: Batch Synthesis
+        with gr.Tab("Batch (Long Text)"):
+            gr.Markdown("### For texts over 2000 characters")
             gr.Markdown("Automatically splits into chunks and combines audio.")
             
             with gr.Row():
                 with gr.Column(scale=1):
-                    long_text = gr.Textbox(
-                        label="Long Text",
-                        lines=12,
-                        max_length=10000,
-                        placeholder="Paste long text here (up to 10000 characters)..."
-                    )
+                    long_text = gr.Textbox(label="Long Text", lines=12, max_length=10000, placeholder="Paste long text here (up to 10000 characters)...")
                     chunk_slider = gr.Slider(300, 800, value=500, step=50, label="Chunk Size (characters)")
                     batch_audio = gr.Audio(label="Reference Voice", type="filepath", sources=["upload", "microphone"])
                     with gr.Row():
                         batch_lang = gr.Dropdown(choices=SUPPORTED_LANGUAGES, value="en", label="Language")
                         batch_speed = gr.Slider(0.5, 2.0, value=1.0, step=0.05, label="Speed")
-                    batch_btn = gr.Button("🎵 Generate Long Speech", variant="primary", size="lg")
+                    batch_btn = gr.Button("Generate Long Speech", variant="primary", size="lg")
                 with gr.Column(scale=1):
                     batch_output = gr.Audio(label="Generated Speech (Combined)", type="filepath")
                     
-                    gr.Markdown("""
-                    ### ℹ️ Batch Mode Info
-                    - Splits at sentence boundaries (., !, ?)
-                    - Synthesizes each chunk separately
-                    - Combines all chunks into one file
-                    - Best for audiobooks, presentations, long narrations
-                    """)
+                    gr.Markdown("### Batch Mode Info")
+                    gr.Markdown("- Splits at sentence boundaries (., !, ?)")
+                    gr.Markdown("- Synthesizes each chunk separately")
+                    gr.Markdown("- Combines all chunks into one file")
+                    gr.Markdown("- Best for audiobooks, presentations, long narrations")
             
             batch_btn.click(batch_synthesize, [long_text, batch_audio, batch_lang, batch_speed, chunk_slider], [batch_output])
         
-        # ========================================================
-        # TAB 3: HELP & TROUBLESHOOTING
-        # ========================================================
-        with gr.Tab("❓ Help"):
-            gr.Markdown("""
-            # 📖 Help Guide
-            
-            ## ✅ Quick Start
-            
-            1. **Record your voice** (click microphone icon, say 6-10 seconds)
-            2. **Type text** you want to synthesize
-            3. **Click Generate** - works in 10-30 seconds
-            
-            ## 🔧 Troubleshooting
-            
-            ### "Index out of range" / Voice extraction fails
-            - **Use microphone recording** - this works 100% of the time
-            - Uploaded files must be short (6-10 seconds), WAV format, clean speech
-            
-            ### First use is slow
-            - Normal! Downloads 4GB model (2-5 minutes)
-            - Subsequent uses are fast (10-30 seconds)
-            - Enable GPU in Space settings for faster performance
-            
-            ### Improve quality
-            - Use 8-10 second recording
-            - No background noise
-            - One speaker only
-            - Speak clearly at normal pace
-            
-            ## 🔄 Converting your existing audio
-            
-            ```bash
-            # Extract best 10 seconds from long audio
-            ffmpeg -i long_audio.mp3 -t 10 -ac 1 -ar 22050 short.wav
-            
-            # Remove silence
-            ffmpeg -i input.wav -af silenceremove=1:0:-50dB output.wav
+        # TAB 3: Help
+        with gr.Tab("Help & Troubleshooting"):
+            gr.Markdown("# Help Guide")
+            gr.Markdown("")
+            gr.Markdown("## Quick Start")
+            gr.Markdown("")
+            gr.Markdown("1. Record your voice (click microphone icon, say 6-10 seconds)")
+            gr.Markdown("2. Type text you want to synthesize")
+            gr.Markdown("3. Click Generate - works in 10-30 seconds")
+            gr.Markdown("")
+            gr.Markdown("## Troubleshooting")
+            gr.Markdown("")
+            gr.Markdown("### Voice extraction fails")
+            gr.Markdown("- Use microphone recording - this works 100% of the time")
+            gr.Markdown("- Uploaded files must be short (6-10 seconds), WAV format, clean speech")
+            gr.Markdown("")
+            gr.Markdown("### First use is slow")
+            gr.Markdown("- Normal! Downloads 4GB model (2-5 minutes)")
+            gr.Markdown("- Subsequent uses are fast (10-30 seconds)")
+            gr.Markdown("- Enable GPU in Space settings for faster performance")
+            gr.Markdown("")
+            gr.Markdown("### Improve quality")
+            gr.Markdown("- Use 8-10 second recording")
+            gr.Markdown("- No background noise")
+            gr.Markdown("- One speaker only")
+            gr.Markdown("- Speak clearly at normal pace")
+            gr.Markdown("")
+            gr.Markdown("## Converting your existing audio")
+            gr.Markdown("")
+            gr.Markdown("```bash")
+            gr.Markdown("# Extract best 10 seconds from long audio")
+            gr.Markdown("ffmpeg -i long_audio.mp3 -t 10 -ac 1 -ar 22050 short.wav")
+            gr.Markdown("# Remove silence")
+            gr.Markdown("ffmpeg -i input.wav -af silenceremove=1:0:-50dB output.wav")
+            gr.Markdown("```")
+            gr.Markdown("")
+            gr.Markdown("## Need more help?")
+            gr.Markdown("Check container logs in Space Settings.")
+
+if __name__ == "__main__":
+    demo.launch(
+        server_name="0.0.0.0",
+        server_port=int(os.environ.get("PORT", 7860)),
+        show_error=True
+    )