Update app.py

app.py

@@ -39,10 +39,10 @@ def predict(image, text):
     return response
 
 # Whisper STT optional model
-@spaces.GPU
-def transcribe_audio(audio):
-    result = whisper.transcribe(audio, model="base")
-    return result["text"]
+#@spaces.GPU
+#def transcribe_audio(audio):
+#    result = whisper.transcribe(audio, model="base")
+#    return result["text"]
 
 # Example photos and prompts
 example_images = [