Upload folder using huggingface_hub

app.py

@@ -1,8 +1,7 @@
 import gradio as gr
 import numpy as np
 from PIL import Image
-import tempfile
-import os
+import time
 
 # Configuration
 MAX_SEED = np.iinfo(np.int32).max
@@ -10,14 +9,13 @@ MAX_SEED = np.iinfo(np.int32).max
 def generate_video(
     image,
     audio,
-    prompt="A person talking",
-    resolution="480p",
-    audio_cfg=2.5,
-    guidance_scale=5.0,
-    num_inference_steps=25,
-    seed=42,
-    max_duration=10,
-    progress=gr.Progress()
+    prompt,
+    resolution,
+    audio_cfg,
+    guidance_scale,
+    num_inference_steps,
+    seed,
+    max_duration
 ):
     """Generate talking video from image and audio"""
     
@@ -28,19 +26,9 @@ def generate_video(
         return None, "❌ Please upload an audio file"
     
     try:
-        progress(0, "Initializing...")
-        
-        # For now, return a placeholder message since we need to implement the actual model
-        # In a real implementation, you would load the MeiGen-MultiTalk model here
-        
-        progress(0.5, "Processing audio and image...")
-        
         # Simulate processing time
-        import time
         time.sleep(2)
         
-        progress(1.0, "Video generation complete!")
-        
         return None, f"""✅ Video generation request processed!
 
 **Settings:**
@@ -66,129 +54,103 @@ The model files are not included in this demo due to size constraints."""
 def randomize_seed():
     return np.random.randint(0, MAX_SEED)
 
-# Gradio Interface
-with gr.Blocks(
-    theme=gr.themes.Soft(),
-    title="MeiGen-MultiTalk Demo",
-    css="""
-    .main-header {
-        text-align: center;
-        background: linear-gradient(45deg, #ff6b6b, #4ecdc4);
-        -webkit-background-clip: text;
-        -webkit-text-fill-color: transparent;
-        background-clip: text;
-        font-size: 2.5em;
-        font-weight: bold;
-        margin-bottom: 0.5em;
-    }
-    .subtitle {
-        text-align: center;
-        color: #666;
-        margin-bottom: 2em;
-    }
-    """
-) as demo:
+# Simplified Gradio Interface
+with gr.Blocks(title="MeiGen-MultiTalk Demo") as demo:
     
     gr.HTML("""
-    <div class="main-header">🎬 MeiGen-MultiTalk Demo</div>
-    <p class="subtitle">Generate talking videos from images and audio using AI</p>
+    <div style="text-align: center; margin-bottom: 20px;">
+        <h1>🎬 MeiGen-MultiTalk Demo</h1>
+        <p>Generate talking videos from images and audio using AI</p>
+    </div>
     """)
     
     with gr.Row():
         # Input Column
-        with gr.Column(scale=1):
+        with gr.Column():
             gr.Markdown("### 📁 Input Files")
             
             image_input = gr.Image(
                 label="Reference Image",
-                type="pil",
-                height=300
+                type="pil"
             )
             
             audio_input = gr.Audio(
-                label="Audio File",
-                type="filepath"
+                label="Audio File"
             )
             
             prompt_input = gr.Textbox(
                 label="Prompt",
                 placeholder="A person talking naturally...",
-                value="A person talking",
-                lines=2
+                value="A person talking"
             )
             
             gr.Markdown("### ⚙️ Generation Settings")
             
-            with gr.Row():
-                resolution = gr.Dropdown(
-                    choices=["480p", "720p"],
-                    value="480p",
-                    label="Resolution"
-                )
-                
-                max_duration = gr.Slider(
-                    minimum=1,
-                    maximum=15,
-                    value=10,
-                    step=1,
-                    label="Max Duration (seconds)"
-                )
+            resolution = gr.Dropdown(
+                choices=["480p", "720p"],
+                value="480p",
+                label="Resolution"
+            )
             
-            with gr.Row():
-                audio_cfg = gr.Slider(
-                    minimum=1.0,
-                    maximum=5.0,
-                    value=2.5,
-                    step=0.1,
-                    label="Audio CFG Scale"
-                )
-                
-                guidance_scale = gr.Slider(
-                    minimum=1.0,
-                    maximum=10.0,
-                    value=5.0,
-                    step=0.5,
-                    label="Guidance Scale"
-                )
+            max_duration = gr.Slider(
+                minimum=1,
+                maximum=15,
+                value=10,
+                step=1,
+                label="Max Duration (seconds)"
+            )
             
-            with gr.Row():
-                num_inference_steps = gr.Slider(
-                    minimum=10,
-                    maximum=50,
-                    value=25,
-                    step=1,
-                    label="Inference Steps"
-                )
-                
-                seed = gr.Number(
-                    value=42,
-                    minimum=0,
-                    maximum=MAX_SEED,
-                    label="Seed"
-                )
+            audio_cfg = gr.Slider(
+                minimum=1.0,
+                maximum=5.0,
+                value=2.5,
+                step=0.1,
+                label="Audio CFG Scale"
+            )
+            
+            guidance_scale = gr.Slider(
+                minimum=1.0,
+                maximum=10.0,
+                value=5.0,
+                step=0.5,
+                label="Guidance Scale"
+            )
+            
+            num_inference_steps = gr.Slider(
+                minimum=10,
+                maximum=50,
+                value=25,
+                step=1,
+                label="Inference Steps"
+            )
+            
+            seed = gr.Number(
+                value=42,
+                minimum=0,
+                maximum=MAX_SEED,
+                label="Seed"
+            )
             
             with gr.Row():
-                randomize_btn = gr.Button("🎲 Randomize Seed", variant="secondary")
-                generate_btn = gr.Button("🎬 Generate Video", variant="primary", size="lg")
+                randomize_btn = gr.Button("🎲 Randomize Seed")
+                generate_btn = gr.Button("🎬 Generate Video", variant="primary")
         
         # Output Column
-        with gr.Column(scale=1):
+        with gr.Column():
             gr.Markdown("### 🎥 Generated Video")
             
             video_output = gr.Video(
-                label="Generated Video",
-                height=400
+                label="Generated Video"
             )
             
             result_text = gr.Textbox(
                 label="Generation Log",
-                lines=8,
-                max_lines=15
+                lines=10
             )
     
-    # Examples
-    gr.Markdown("### 📋 Tips for Best Results")
+    # Tips
     gr.Markdown("""
+    ### 📋 Tips for Best Results
     - **Image**: Use clear, front-facing photos with good lighting
     - **Audio**: Ensure clean audio without background noise
     - **Prompt**: Be specific about the desired talking style
@@ -219,10 +181,6 @@ with gr.Blocks(
     )
 
 if __name__ == "__main__":
-    demo.launch(
-        share=False,
-        server_port=7860,
-        show_error=True
-    )
+    demo.launch(share=True)
 
 # Force restart - updated at 2025-01-23

requirements.txt

@@ -1,4 +1,4 @@
-gradio==4.44.1
+gradio==4.36.1
 torch
 pillow
 numpy