Quickstart - Teams Voice Plugin (Stand in)

Pick your runtime. Both paths get you to a live Teams call where your agent answers in voice.

Prerequisites

A working OpenClaw or Hermes Agent install (the host).
A media worker that bridges the live Teams call into the plugin over an HMAC WebSocket (the worker renders the avatar tile and forwards audio/video; it dials this plugin).
A realtime voice provider key (for realtime mode) - e.g. OpenAI or Azure OpenAI. Streaming mode uses your host’s configured STT/TTS instead.

OpenClaw (npm)
Hermes (pip)

Install

openclaw plugins install clawhub:@komaa/msteams-voice
cd extensions/msteams-voice && pnpm install && pnpm build

Configure

Add a msteams-voice entry to your OpenClaw config - sharedSecret must match the worker:

{
  "plugins": {
    "entries": {
      "msteams-voice": {
        "config": {
          "enabled": true,
          "mode": "realtime",
          "port": 9442,
          "path": "/voice/msteams/stream",
          "sharedSecret": "<same secret as the Teams worker>",
          "inboundPolicy": "allowlist",
          "allowFrom": ["<caller AAD object id>"],
          "realtime": {
            "provider": "openai",
            "providers": {
              "openai": { "apiKey": "<key>", "model": "gpt-realtime" }
            }
          }
        }
      }
    }
  }
}

Full reference → OpenClaw configuration.

Run + call

Start the gateway (openclaw gateway run), point the worker’s WebSocket URL at ws://<host>:9442/voice/msteams/stream with the matching secret, then place a Teams call to the bot.

Install

Into the same Python environment as Hermes:

uv pip install --python /path/to/hermes/venv/bin/python hermes-plugin-teams-voice
hermes plugins enable teams_voice

Configure

In ~/.hermes/config.yaml under plugins.entries.teams_voice.config (secrets in .env):

plugins:
  enabled: [teams_voice]
  entries:
    teams_voice:
      config:
        shared_secret: ${TEAMS_VOICE_SHARED_SECRET}   # must match the worker
        host: 127.0.0.1
        port: 8443
        realtime:
          backend: azure
          azure_endpoint: https://<your-azure-resource>.cognitiveservices.azure.com
          azure_deployment: gpt-realtime
          api_key: ${AZURE_FOUNDRY_API_KEY}

Full reference → Hermes configuration.

Run + call

hermes teams-voice serve --handler realtime

Point the worker’s WebSocket URL at ws://127.0.0.1:8443/voice/msteams/stream with the matching secret, then place a Teams call to the bot.

Streaming mode (no realtime key needed) is a one-line change - see Modes.

​Prerequisites

Prerequisites