### Answer Retell AI has separate agent types for voice and chat interactions. Voice agents handle phone calls using speech recognition, LLMs, and text-to-speech [(1) ]() , while Chat Agents are a distinct entity with their own configuration and API endpoints [(2) ]() . When you create and deploy an agent, you're explicitly building either a Voice Agent or a Chat Agent—they are separate products within the platform. The Chat Agent has its own specific settings like

auto_close_message

and

end_chat_after_silence_ms

that are tailored for text-based conversations . So the AI doesn't need to "recognize" which type of conversation it's in at runtime—the channel is determined by which agent type you've deployed and how the conversation is initiated (phone call vs. chat widget/API) .

​ To learn more, see the sources I considered: [[1] AI Voice Agent](https://www.retellai.com/glossary/ai-voice-agent) [[2] Update Chat Agent](https://docs.retellai.com/api-references/update-chat-agent) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.