Update app.py

app.py

@@ -378,12 +378,12 @@ def vqa_answer(image, question):
     answer = vqa_processor.decode(out[0], skip_special_tokens=True)
 
     if not is_caption_safe(answer):
-    beep = make_beep_sound()
-    tts = gTTS("Warning! Unsafe or inappropriate content detected.", lang="en")
-    speech_tmp = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3")
-    tts.save(speech_tmp.name)
-    combined = combine_audio(beep, speech_tmp.name)
-    return "⚠️ Warning: Unsafe or inappropriate content detected!", combined
+        beep = make_beep_sound()
+        tts = gTTS("Warning! Unsafe or inappropriate content detected.", lang="en")
+        speech_tmp = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3")
+        tts.save(speech_tmp.name)
+        combined = combine_audio(beep, speech_tmp.name)
+        return "⚠️ Warning: Unsafe or inappropriate content detected!", combined
 
 
     return answer, None