Chamade joins meetings, phone calls, and DMs across 10+ platforms. Drop your ElevenLabs or Deepgram key in the dashboard — we run the STT/TTS pipeline for you, free on our side. Prefer full control? The raw-PCM audio WebSocket is always there, plug in OpenAI Realtime, LiveKit, Pipecat, whatever. Drive the call from any MCP client, REST API, or WebSocket. One gateway for Teams, Meet, Discord, Zoom, Telegram, Slack, WhatsApp, Nextcloud Talk, and phone.
One uniform API for voice and chat across 10+ platforms. We handle the transport, the hosted STT/TTS pipeline (you bring your ElevenLabs / Deepgram key — free on our side), and the control plane for chat, DMs, typing indicators. Want full control over the voice models? The raw-PCM audio WebSocket is always there too.
Join meetings via Graph API. Raw PCM audio bridge + meeting chat relay.
Voice ChatNative SDK integration. Raw PCM audio bridge + meeting chat relay.
Voice ChatWebRTC connection. Inbound audio only (Meet Media API, Developer Preview) + in-meeting chat.
Voice (listen) ChatVoice channel join via bot. Per-user audio capture + text channel chat.
Voice ChatDMs and group chats via Bot API. Shared bot or bring your own.
ChatInvite link to chat with your agent. No app install needed.
ChatDMs and channels. Install to workspace via OAuth.
ChatJoin Talk rooms via addon. Raw PCM audio bridge + chat relay.
Voice ChatCall any phone number worldwide. Raw 8 kHz PCM bridge, bring your own SIP trunk and DIDs.
VoiceDrop your ElevenLabs or Deepgram key in the dashboard once. Chamade joins the meeting, runs the full speech pipeline on its side with your key, and your agent only ever sees text — transcripts flow in as events, a single tool call speaks into the meeting. No Chamade charge, you only pay your provider. Already have a voice stack (OpenAI Realtime, LiveKit, Pipecat)? The raw-PCM audio WebSocket is still there for when you want full control.
Connect your accounts on the dashboard. Add the MCP server or call the REST API. Done.
Sign in to your Chamade dashboard and link the platforms you use. One OAuth click per service — we store the credentials securely, your agent never sees them.
Each platform opens its standard OAuth consent screen. You approve the permissions Chamade needs — calendar access to join meetings, audio streaming rights. No passwords stored, tokens are encrypted at rest.
chamade.io is requesting the following permissions:
Your key is ready. It carries your connected platforms with it — any agent that uses this key can access all your authorized services. Rotate or revoke anytime from the dashboard.
chmd_live_a8f3e2b1c9d4...
One URL and your API key. Drop this into Claude Desktop, Claude Code, Cursor, Windsurf, or any MCP-compatible client that speaks the Streamable HTTP transport. Nothing to install, no local server to keep running. Or call the REST API directly from any language.
One uniform layer under every voice and chat use case. We own the platform quirks and the hosted STT/TTS pipeline (BYOK on the provider side — you pick ElevenLabs, Deepgram, …). If you want to run the speech models yourself, the raw-PCM WebSocket is one call away. MCP server, REST API, or WebSocket — take your pick.
Tell Claude "join my 2pm standup". Chamade bridges the meeting, streams transcripts into the agent, and speaks back through hosted TTS using your ElevenLabs key. Works with Claude Desktop, Claude Code, Cursor, Windsurf, or any MCP client. Swap to BYO audio on the WebSocket if you prefer your own pipeline.
Build a phone agent, an IVR, or a smart answering machine. Chamade picks up the SIP call and runs hosted STT/TTS with your provider key — or hands you raw 8 kHz PCM on a WebSocket if you want to plug in OpenAI Realtime, Deepgram Voice Agent, or your own cascade. No telephony stack needed on your side.
Half your team is on Teams, the other on Discord? Bridge them with audio and chat relayed both ways. Your application sits in the middle and can moderate, translate, or log.
Already running a LiveKit Agent, a Pipecat bot, or a Vocode pipeline? Chamade is the missing piece that turns "localhost WebRTC" into Teams, Meet, Zoom, Discord, NC Talk, and PSTN — same PCM contract, one API key.
No subscription, no quota, no billing. Pricing will be announced once the feature surface stabilizes — we'll give early-access users reasonable notice.
For everyone, right now
Need to talk production-scale, SLAs or commercial usage? [email protected]
Create your account, connect your platforms, and get an API key in under a minute.
Create your account