Update app.py

app.py

@@ -7,6 +7,7 @@ import gradio as gr
 import torch
 from PIL import Image
 import time
+import spaces
 
 # Import pipeline modules
 from utils.transcribe import transcribe_audio, list_available_whisper_models
@@ -70,6 +71,7 @@ DEFAULT_STYLE_SUFFIX = "cinematic, 35 mm, shallow depth of field, film grain"
 IMAGE_MODES = ["Independent", "Consistent (Img2Img)"]
 DEFAULT_IMAGE_MODE = "Independent"
 
+@spaces.GPU
 def process_audio(
     audio_path, 
     whisper_model,