@@ -64,14 +64,14 @@ async def entrypoint(ctx: JobContext):
6464 # Set up a voice AI pipeline using OpenAI, Cartesia, Deepgram, and the LiveKit turn detector
6565 session = AgentSession (
6666 # A Large Language Model (LLM) is your agent's brain, processing user input and generating a response
67- # See all providers at https://docs.livekit.io/agents/integrations /llm/
67+ # See all available models at https://docs.livekit.io/agents/models /llm/
6868 llm = "openai/gpt-4o-mini" ,
6969 # Speech-to-text (STT) is your agent's ears, turning the user's speech into text that the LLM can understand
70- # See all providers at https://docs.livekit.io/agents/integrations /stt/
70+ # See all available models at https://docs.livekit.io/agents/models /stt/
7171 stt = "deepgram/nova-3" ,
7272 # Text-to-speech (TTS) is your agent's voice, turning the LLM's text into speech that the user can hear
73- # See all providers at https://docs.livekit.io/agents/integrations /tts/
74- tts = "cartesia:6f84f4b8-58a2-430c-8c79-688dad597532" ,
73+ # See all available models as well as voice selections at https://docs.livekit.io/agents/models /tts/
74+ tts = "cartesia/sonic-2 :6f84f4b8-58a2-430c-8c79-688dad597532" ,
7575 # VAD and turn detection are used to determine when the user is speaking and when the agent should respond
7676 # See more at https://docs.livekit.io/agents/build/turns
7777 # turn_detection=MultilingualModel(),
@@ -82,7 +82,7 @@ async def entrypoint(ctx: JobContext):
8282 )
8383
8484 # To use a realtime model instead of a voice pipeline, use the following session setup instead.
85- # This is for OpenAI Realtime API, for other providers, see https://docs.livekit.io/agents/integrations /realtime/
85+ # (Note: This is for the OpenAI Realtime API. For other providers, see https://docs.livekit.io/agents/models /realtime/))
8686 # 1. Install livekit-agents[openai]
8787 # 2. Set OPENAI_API_KEY in .env.local
8888 # 3. Add `from livekit.plugins import openai` to the top of this file
0 commit comments