Vexavexa
Book a demo
n8n · Workflow automation · Apache 2.0

n8n
Meeting
Transcription

Vexa has no n8n node: an HTTP Request node sends the bot, and another fetches the transcript when the meeting ends.

Get an API keyView on GitHub
2.9k stars on GitHub
vexa · api
Zoom · Teams · MeetVexa botTranscriptsYour workflowREST · webhooks
  • Zoom · Teams · Meet
  • Vexa bot
  • Transcripts
  • Your workflow
  • REST · webhooks
Teams
Meet
Vexa
🦀
Meeting data for your agents

Your agents read the meeting while it runs.

The transcript streams to your agents as people speak, over WebSocket or MCP.

Built for n8n

Vexa supplies two plain HTTP calls, send a bot and fetch the transcript, and n8n does the rest.

  • HTTP Request nodes, with your API key as a credential
  • Meet, Teams and Zoom through the same two calls
  • Self-host n8n and Vexa; the transcript stays in your network
n8n HTTP Request node 1: send the bot
{
  "method": "POST",
  "url": "{{$env.VEXA_API_BASE}}/bots",
  "headers": { "X-API-Key": "{{$env.VEXA_API_KEY}}" },
  "body": {
    "platform": "google_meet",
    "native_meeting_id": "{{$json.meeting_code}}",
    "bot_name": "Vexa"
  }
}

// Node 2, when the meeting ends: fetch the transcript
//   GET {{$env.VEXA_API_BASE}}/transcripts/google_meet/{{$json.meeting_code}}
// Start it from Vexa's meeting.completed webhook arriving at an
// n8n Webhook node, or from a Wait node.
Set $API_BASE to https://api.cloud.vexa.ai (hosted) or http://localhost:18056 (self-host). Get $API_KEY from /start.

Common patterns

What you can build with Vexa and n8n

Three workflows, each with two Vexa calls.

Summaries in Slack

Your team wants a summary in Slack after each call, without writing a backend.

What you get
  • Calendar trigger, send the bot, fetch the transcript.
  • Summarize it with your model and post it to Slack.
  • This is the community template's path.

Transcripts into your notes

Each customer call should land in Notion or Drive, by speaker.

What you get
  • When the webhook arrives, fetch the transcript.
  • Create one page per meeting.

Into your CRM

Your team wants each call's notes on the account in the CRM.

What you get
  • Fetch the transcript and summarize it.
  • Write it with n8n's CRM node.
  • Vexa does not connect to the CRM itself.

Wire Vexa into n8n

No custom node. Two HTTP Request nodes and your API key.

Quickstart

  1. Get a Vexa API keySign in at vexa.ai. Self-hosted: make all prints a key.
  2. Store the key in n8nAdd it as a credential sent as the X-API-Key header.
  3. Send the botPOST /bots with the platform and the meeting ID.
  4. Fetch the transcriptWhen the meeting ends, GET /transcripts/{platform}/{native_meeting_id}.
  5. Send it onSlack, Notion, Drive, a model: any node n8n has.
Get an API keyRead the n8n guide

n8n architecture

n8n orchestrates; Vexa joins the meeting and makes the transcript.

Pipeline
bot → audio capture → speech-to-text → transcript segments → REST / WebSocket

Before you start

  • The community template uses an old base URL; update it.
  • Vexa's webhook says a meeting ended; fetch the transcript next.

Documentation

Go deeper on n8n

The template, webhooks and calendar triggers.

Full reference at docs.vexa.ai.

Related reading

Keep going on n8n

Related pages, guides and comparisons.

n8n FAQ

What automation builders ask before they wire Vexa into n8n.

Is there an n8n node for Vexa?

No. Vexa works through n8n's HTTP Request node: one call sends the bot, one fetches the transcript. A community template on n8n.io wires a Google Meet call to a Slack summary.

Can the workflow start from a calendar invite?

Yes. A Google Calendar or Outlook trigger, a filter for the meetings you want, then the HTTP Request that sends the bot. Vexa can also join the meetings in a connected calendar by itself, without n8n.

How does the transcript come back into n8n?

Two ways. Vexa sends a signed meeting.completed webhook when the meeting ends; point it at an n8n Webhook node and fetch the transcript from there. Or wait, then call GET /transcripts/{platform}/{native_meeting_id}.

Can both run inside our network?

Yes. n8n and Vexa both self-host. With Vexa's transcription unit and your model endpoint inside the network too, the meeting audio and transcript don't leave it.

A seat, an hour, or free

A seat for your agent. An hour for what you build. No licence fee when you run it yourself.

Pricing

Build the workflow. Two calls, then anything n8n reaches.

Sign in, copy an API key, add two HTTP Request nodes. Free credit, no card.

Get an API keyn8n guideGitHub

Apache 2.0 · Self-host or cloud · No credit card required