Update app.py

app.py

@@ -68,7 +68,7 @@ def audioMode(audio):
     input_features = processor_audio(audio_array, sampling_rate=16000, return_tensors="pt").input_features 
     predicted_ids = model_audio.generate(input_features)
     transcription = processor_audio.batch_decode(predicted_ids, skip_special_tokens=True)
-    return transcription
+    return transcription[0]
 
 
 interface_title = "TSAI-ERA-V1 - Capstone - Multimodal GPT Demo"