Are azure open ai/aws bedrock llms supported on livekit self deployed?

Thanks for reaching out, harshal. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.

Azure open ai

Thanks — I'll show how to use Azure OpenAI with a self‑deployed LiveKit via the LiveKit Agents framework and give concise example code (Python and Node.js). Key points • Use the Azure OpenAI plugin for LiveKit Agents. • Required: Azure endpoint (e.g., https://<your-resource>.openai.azure.com/), the deployment name, and an API key or Microsoft Entra ID token. • This works on self‑hosted LiveKit; LiveKit Cloud is not required. Python example

from livekit.agents import AgentSession
from livekit.plugins import openai

session = AgentSession(
    llm=openai.LLM.with_azure(
        azure_deployment="<your-deployment-name>",
        azure_endpoint="https://<your-resource>.<http://openai.azure.com/|openai.azure.com/>",
        api_key="<your-api-key>",
        api_version="2024-10-01-preview",
    )
)

# use `session` to run agent tasks, prompts, etc.

Node.js example

import { AgentSession } from "livekit-agents";
import { openai } from "livekit-plugins";

const session = new AgentSession({
  llm: openai.LLM.withAzure({
    azure_deployment: "<your-deployment-name>",
    azure_endpoint: "https://<your-resource>.<http://openai.azure.com/|openai.azure.com/>",
    api_key: "<your-api-key>",
    api_version: "2024-10-01-preview",
  }),
});

// use `session` to run agent tasks, prompts, etc.

Next steps I can help with • Sample prompt/agent flow for your use case (moderation, transcription, summarization, etc.) • How to provide credentials securely on your self‑hosted server • Debug steps if you hit authentication or network errors Which would you like next?

I can use these as Voice Agent LLMs right?

Yes — you can use Azure OpenAI models as the LLM for Voice Agents in a self‑deployed LiveKit via the LiveKit Agents framework. What to know and next steps: • Use the Azure OpenAI plugin and configure

azure_endpoint

,

azure_deployment

, and credentials (API key or Microsoft Entra token). • Ensure your Voice Agent flow sends audio (or transcriptions) to the LLM as text prompts and receives text responses to convert to speech. LiveKit Agents handles the LLM integration; text-to-speech (TTS) must be implemented separately (you can use Azure TTS or another TTS service). • Securely provide credentials on your self-hosted server (environment variables or secret store). Would you like: • a concrete example that includes audio → transcription → LLM prompt → TTS, or • a minimal code snippet showing how to wire Azure OpenAI into a Voice Agent session?