Skip to main content
@komaa/deepgram-msteams-bridge brings a Deepgram Voice Agent (Nova STT + your chosen LLM + Aura TTS, with turn-taking, all run by Deepgram) onto a Microsoft Teams call. It is the Deepgram analogue of the ElevenLabs, LiveKit and OpenAI bridges: it hosts the HMAC WebSocket that StandIn connects to, then relays audio to and from a Deepgram Voice Agent session.
Like the other bridges, this is a standalone Node service, not a framework plugin. There is also no agent to configure in a dashboard: the bridge configures each Voice Agent session itself (STT model, LLM, voice, prompt, greeting, tools) from environment variables. Two variables and it runs.
Prefer Python? The same bridge exists as a Python package: deepgram-msteams-bridge on PyPI (pip install deepgram-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 opens one Voice Agent WebSocket per call (wss://agent.deepgram.com/v1/agent/converse), waits for the server’s Welcome, sends a Settings message configuring the agent, and relays audio both ways. Audio is copy-only: the StandIn wire is base64 PCM 16 kHz and the Voice Agent session is pinned to linear16 at 16 kHz, so nothing is resampled or re-encoded. Caller barge-in (UserStartedSpeaking) maps to a Teams-side flush, and stale “ghost” audio is dropped so nothing plays after the caller cuts in. The agent gets four built-in tools automatically (end_call, look for vision, show_image, express for avatar emotion), and you can register your own function tools the bridge executes in-process. See Extending the agent’s tools.

​
Requirements

  • Node.js >= 20.
  • A Deepgram API key with Voice Agent access (console.deepgram.com).
  • 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: 
DEEPGRAM_API_KEY=dg_... \
WORKER_SHARED_SECRET=... \
npx @komaa/deepgram-msteams-bridge
Optionally shape the agent (all have defaults): 
DEEPGRAM_PROMPT="You are Komaa's friendly receptionist. Keep replies short."
DEEPGRAM_GREETING="Hello! You've reached Komaa. How can I help?"
DEEPGRAM_SPEAK_MODEL=aura-2-thalia-en   # Aura voice
DEEPGRAM_THINK_MODEL=gpt-4o-mini        # LLM
Or add it to a project (npm i @komaa/deepgram-msteams-bridge) and embed it - see the library API for the programmatic surface, custom tools, and a custom vision hook. The bridge binds its media WebSocket (default ws://<host>:8080/voice/msteams/stream; StandIn appends /{callId} per call).
Third-party LLMs need an endpoint. Deepgram-managed open_ai and anthropic work with just a model name. Providers like google, groq or aws_bedrock additionally require DEEPGRAM_THINK_ENDPOINT_URL (Deepgram dials it) - see the configuration reference.

​
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 Deepgram 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 two environment variables, expose the port, connect it to StandIn, and place a call - with the vision hook and a custom lookup_order tool explained along the way.
Next: the full configuration reference - every environment variable, the call governor, vision, and BYO-LLM endpoints. Deep protocol and library docs live on the project site.

Docs site

npm

Source