Update app.py

app.py

@@ -1,42 +1,62 @@
 import gradio as gr
+import torch
 import subprocess
-from pathlib import Path
 import os
+from pathlib import Path
 
-def generate(image, audio):
-    if not image or not audio:
-        return None, "Upload both files"
-    
+def setup():
+    """One-time setup"""
+    if not Path("DiffSynth-Studio").exists():
+        subprocess.run("git clone https://github.com/modelscope/DiffSynth-Studio.git", shell=True)
+        subprocess.run("pip install -e ./DiffSynth-Studio", shell=True)
+
+def generate_video(image, audio):
     try:
-        # Use makeittalk - simple and reliable
-        result = subprocess.run([
-            "python", "main_end2end.py",
-            "--jpg", image,
-            "--audio", audio,
-            "--config", "config/simple.yaml"
-        ], capture_output=True, cwd="/app/makeittalk")
+        if not image or not audio:
+            return None, "Please upload both image and audio!"
+        
+        setup()
+        
+        # Run inference
+        output_path = "output.mp4"
         
-        output = Path("/app/makeittalk/examples/output.mp4")
-        return str(output) if output.exists() else None, "Done!"
+        cmd = f"""
+cd DiffSynth-Studio && python examples/video_generation/musetalk.py \
+--image_path ../{image} \
+--audio_path ../{audio} \
+--output_path ../{output_path}
+"""
+        
+        result = subprocess.run(cmd, shell=True, capture_output=True, text=True)
+        
+        if Path(output_path).exists():
+            return output_path, "✅ Video generated successfully!"
+        else:
+            return None, f"❌ Failed: {result.stderr}"
+            
     except Exception as e:
-        return None, str(e)
-
-demo = gr.Interface(
-    fn=generate,
-    inputs=[
-        gr.Image(type="filepath", label="Face"),
-        gr.Audio(type="filepath", label="Audio")
-    ],
-    outputs=[
-        gr.Video(label="Result"),
-        gr.Textbox(label="Status")
-    ],
-    title="AI Lip Sync"
-)
+        return None, f"❌ Error: {str(e)}"
 
-# Setup on start
-if not Path("/app/makeittalk").exists():
-    subprocess.run(["git", "clone", "https://github.com/yzhou359/MakeItTalk.git", "/app/makeittalk"])
-    subprocess.run(["bash", "quick_demo.sh"], cwd="/app/makeittalk")
+# Gradio Interface
+with gr.Blocks(title="AI Lip Sync") as app:
+    gr.Markdown("# 🎤 AI Lip Sync Generator")
+    gr.Markdown("Upload a face image and audio to generate lip-synced video")
+    
+    with gr.Row():
+        with gr.Column():
+            image_input = gr.Image(type="filepath", label="📷 Face Image")
+            audio_input = gr.Audio(type="filepath", label="🎵 Audio File")
+            generate_btn = gr.Button("🚀 Generate Video", variant="primary")
+        
+        with gr.Column():
+            video_output = gr.Video(label="📹 Generated Video")
+            status_output = gr.Textbox(label="Status", lines=2)
+    
+    generate_btn.click(
+        fn=generate_video,
+        inputs=[image_input, audio_input],
+        outputs=[video_output, status_output]
+    )
 
-demo.launch()
+if __name__ == "__main__":
+    app.launch()