Update app.py

app.py

@@ -74,7 +74,8 @@ def get_output_video(text):
             dtype=torch.float16 if fp16 else torch.float32,
             device=device
         )
-
+    
+        # Ensure correct dtype for inputs
         image = model.generate_image(
             text,
             seed,
@@ -83,6 +84,7 @@ def get_output_video(text):
             is_verbose=True
         )
         return image
+        
 
     generated_images = []
     for senten in plot[:-1]: