correct stt defect

app.py

@@ -98,7 +98,12 @@ def speech_to_text_loc(audio):
     chunk_length_s=30,
     device=device,
     )
-    text = pipe(audio.copy(), batch_size=2)["text"]
+    print("type of audio:", type(audio))
+    if type(audio) == dict:
+        text = pipe(audio.copy(), batch_size=2)["text"]
+    else:
+        text = pipe(audio, batch_size=2)["text"]
+    
     return text
 
 print("voice to text loc: ", speech_to_text_loc(sample))