session = AgentSession( stt=inference.STT(...
# agents
a
session = AgentSession( stt=inference.STT( model="deepgram/nova-2", language="hi"), llm=inference.LLM(model="openai/gpt-4.1-mini"), tts=inference.TTS( model="cartesia/sonic-3", voice="fd2ada67-c2d9-4afe-b474-6386b87d8fc3" ), turn_detection=MultilingualModel(), vad=ctx.proc.userdata["vad"], # allow the LLM to generate a response while waiting for the end of turn # See more at https://docs.livekit.io/agents/build/audio/#preemptive-generation preemptive_generation=True, ) in this code how to connect it with my api key
r
You can export your keys in the environment and they will be used. Many of plugin readme describe what key is needed: https://github.com/livekit/agents/tree/main/livekit-plugins/livekit-plugins-deepgram#pre-requisites