Skip to main content
@komaa/cartesia-msteams-bridge brings a Cartesia Line agent onto a Microsoft Teams call. It is the Cartesia analogue of the ElevenLabs, LiveKit, OpenAI and Deepgram bridges: it hosts the HMAC WebSocket that StandIn connects to, then relays audio to and from a Line agent stream (Cartesia’s WebSocket API, the integration they provide for bringing your own telephony).
Where the brain lives. A Line agent is your code deployed on Cartesia’s platform - LLM, tools, conversation logic, hangup. This bridge is deliberately a hardened transport: unlike the other bridges there is no in-process tool registry or vision hook, because the Line wire has no client-side tool channel. Your agent still receives the full Teams context (caller identity, participants, active speaker, DTMF, recording state) as metadata and custom events - the contract is documented.
Prefer Python? The same bridge exists as a Python package: cartesia-msteams-bridge on PyPI (pip install cartesia-msteams-bridge), with its own docs site and repo. Same wire protocol, same environment variables, same hardening - pick the runtime that fits your stack. This page follows the Node package.

​
How it works

The bridge mints a short-lived, agent-scoped access token per call (the API key never rides an agent socket), opens wss://api.cartesia.ai/agents/stream/{agentId}, sends the start event, and relays once the server acks. Audio is verbatim: the StandIn wire is base64 PCM 16 kHz and the Line stream is pinned to pcm_16000 with base64 payloads both ways, so the hot path relays the payload string untouched - no decode, no re-encode, no transcoding. Line’s clear (barge-in) maps to a Teams-side playback flush, and keypad digits ride the wire as native dtmf events.

​
Requirements

  • Node.js >= 20.
  • A Cartesia API key and a deployed Line agent (its agent id) - docs.cartesia.ai/line.
  • A StandIn subscription (standin.komaa.com, free package works) - the hosted media bridge that joins the Teams call and connects to this service. See Architecture.
  • Your own Microsoft Teams bot connected to StandIn - the Teams setup walks through the Azure bot, the app package, and the upload. To try it without one, use the sandbox.

​
Run

Env-configured, no install step needed: 
CARTESIA_API_KEY=sk_car_... \
CARTESIA_AGENT_ID=agent_... \
WORKER_SHARED_SECRET=... \
npx @komaa/cartesia-msteams-bridge
Optionally shape the call (your Line agent’s own config is the default): 
CARTESIA_INTRODUCTION="Hello! You've reached Komaa. Quick note: I'm an AI assistant."
CARTESIA_VOICE_ID=a0e99841-438c-4a64-b679-ae501e7d6091   # voice override
Or add it to a project (npm i @komaa/cartesia-msteams-bridge) and embed it - see the library API for the programmatic surface, including the graceful drain() for embedders. The bridge binds its media WebSocket (default ws://<host>:8080/voice/msteams/stream; StandIn appends /{callId} per call).

​
Expose the WebSocket (Agent voice URL)

StandIn connects to the bridge from the internet, so the port must be reachable - a public host or a tunnel. With Tailscale Funnel: 
tailscale funnel --bg --https=8080 8080
Your Agent voice URL is then: 
wss://<machine>.<tailnet>.ts.net:8080/voice/msteams/stream

​
Connect it to StandIn

Register that URL (with a matching shared secret) on your identity in the dashboard, or use the sandbox to try it without your own Teams bot:
  1. Set the identity’s Agent voice URL to where the bridge listens.
  2. Set WORKER_SHARED_SECRET to the shared secret from pairing - both sides must match exactly, or the WebSocket handshake is rejected with 401.
  3. Place a Teams call (or join the sandbox meeting). StandIn joins, connects to the bridge, and your Line agent answers.

​
Try the runnable example

The fastest way to understand the bridge is to run its example project - a minimal, working embedding you can copy straight into your own repo.

Run the example, step by step

Clone examples/basic-bridge, fill in three environment variables, expose the port, connect it to StandIn, and place a call - about 20 lines of code, because the agent’s brain lives in your Line deployment.
Next: the full configuration reference - every environment variable, the call governor, and the Sonic TTS goodbye. Deep protocol and library docs live on the project site, including what your Line agent receives.

Docs site

npm

Source