fix: update tuple audio input handling in transcribe function to extract second element

app.py

@@ -32,7 +32,7 @@ def transcribe(audio, state=""):
     if isinstance(audio, tuple):
         print(f"Tuple contents: {audio}")
         # Try extracting the first element
-        audio = audio[0] if len(audio) > 0 else None
+        audio = audio[1] if len(audio) > 1 else None
         if not audio:
             print("Empty tuple, skipping")
             return state, state