Hello, I'm facing an unusual problem with my agent. Sometimes it sends and say values ​​like |audio_text|, as in the example below: { "id": "item_Clh60LcmD6UIgb8CEL6U4", "type": "message", "role": "assistant", "content": [ "|audio_text|" ], "interrupted": false, "extra": {}, "metrics": { "started_speaking_at": 1765483113.0202394, "stopped_speaking_at": 1765483114.1185155 } } This is my AgentSession: session = AgentSession( userdata=userdata, llm=openai.realtime.RealtimeModel.with_azure(modalities=["text"]), tts=elevenlabs.TTS( voice_settings=VoiceSettings(stability=0.5, similarity_boost=0.75,style=0.0,speed=1.0,use_speaker_boost=True))) Any idea what it could be?

That "<|audio_text|>" (which I find very hard to type!) is coming from and should be internal to the LLM, but sometimes can leak out. Rather than try and adjust the LLM with prompts, it is probably more reliable to intercept and replace these in a custom llm_node (similar to this example, that gets rid of <think> tags)