fix: update tuple audio input handling in transcribe function to extract second element and skip empty tuples

app.py

@@ -33,9 +33,6 @@ def transcribe(audio, state=""):
         print(f"Tuple contents: {audio}")
         # Try extracting the first element
         audio = audio[1] if len(audio) > 1 else None
-        if not audio:
-            print("Empty tuple, skipping")
-            return state, state
         
     if isinstance(audio, np.ndarray):
         audio_buffer.append(audio)