Google Meet
Transcription
API
Real-time Google Meet transcription via open-source REST API. Drop a bot into any Meet, get live transcripts with speaker diarization. Self-host on Docker or Kubernetes, or use Vexa cloud.
Waiting for meeting participants…
Feed meetings into your AI. Live.
AI needs data, and the best data is real-time. Vexa streams transcripts, audio, and context from any meeting straight to your models and agents — as it happens.
Built for Google Meet
Vexa drops a meeting bot into Google Meet calls and streams live transcripts back through a REST API. The bot joins as a regular Meet participant from a headless browser — no Google Workspace add-on, no Gemini API dependency, no per-user licensing. Transcripts are produced by Whisper running locally or in our hosted cloud, so when you self-host, audio never leaves infrastructure you control. The same API also covers Microsoft Teams and Zoom, so one integration handles every platform.
- Join any Meet via the meeting code from the URL — works for Workspace, personal, and external Meet calls
- Live transcript chunks streamed via WebSocket or polled via REST, with speaker diarization
- Apache 2.0 — self-host the bot pod + Whisper service inside your VPC for full data residency
# Extract from https://meet.google.com/abc-defg-hij
# native_meeting_id = "abc-defg-hij"
# (Workspace nickname URLs like /my-team-standup are also supported.)
curl -X POST "$API_BASE/bots" \
-H "Content-Type: application/json" \
-H "X-API-Key: $API_KEY" \
-d '{
"platform": "google_meet",
"native_meeting_id": "abc-defg-hij",
"transcribe_enabled": true,
"transcription_tier": "realtime"
}'
# Poll transcripts
curl "$API_BASE/transcripts/google_meet/abc-defg-hij" \
-H "X-API-Key: $API_KEY"Set $API_BASE to https://api.cloud.vexa.ai (hosted) or http://localhost:8056 (self-host). Get $API_KEY from /get-started.
Google Meet transcription FAQ
Things developers and buyers usually want to know before they wire Google Meet up.
Do I need Google Workspace to use this?
No. Vexa joins Meet calls as a regular participant from a headless browser, just like any guest. It works for personal Meet calls, Workspace calls, and external invites — no Workspace admin setup required. Each bot has its own identity that joins via the meeting URL, the same way you'd invite any external person to a Meet.
How is this different from Google Meet's native captions or Gemini-assisted notes?
Native Meet captions are visual-only — there's no API to pull live transcript chunks programmatically. Gemini-assisted meeting notes are Workspace-tier and asynchronous (notes appear after the meeting in your Drive). Vexa is an API: you POST a meeting URL and receive transcripts as they happen, with millisecond-level chunk timing for downstream automation (CRM updates, real-time agent triggers, MCP tool calls).
Does Vexa record the Meet, or just transcribe?
Vexa is a transcription API, not a recording solution. The bot captures audio for live Whisper transcription and the audio is discarded after transcript chunks are produced — only the text transcript is persisted. If you need audio archives, you can run Vexa next to a separate recording solution, or self-host and modify the audio pipeline.
Can the bot join Workspace meetings restricted to my organization?
If the meeting is set to "your organization only," external participants (including bots) get blocked at the door. Workaround: the meeting organizer invites the bot's email as a guest, or relaxes the join restriction. For meetings inside a single Workspace tenant where Vexa runs alongside the organization's SSO, we have a workspace-trusted deployment pattern — talk to us.
How do I migrate from Recall.AI or another closed-source meeting bot API?
Recall and Vexa share a near-identical REST shape: POST a meeting URL, GET transcripts. Most customers swap the base URL plus rename one or two fields. We document the migration on /compare/recall-ai, including a Recall-API-compatible connector for customers with bigger integration footprints.
Common patterns
What teams build with the Google Meet API
Three shapes we see repeatedly across customers integrating Vexa with Google Meet.
SaaS customer-success calls
Scenario
Your CS team runs onboarding and renewal calls on Google Meet. You want each call summarized, action items extracted, and pushed to the customer record automatically.
What you get
Vexa joins each scheduled call via your CRM webhook, delivers real-time transcripts, and posts a structured summary to HubSpot / Salesforce / Intercom — no manual notes, no missing follow-ups.
User research interviews
Scenario
Research, design, and PM teams run dozens of customer interviews on Meet. You need transcripts that are accurate, searchable, and tagged for synthesis tools like Dovetail or Notion.
What you get
Auto-transcribe every interview, tag turns by speaker, and pipe text into your research platform via the n8n integration. Synthesis cycle drops from days to hours.
Live transcription for AI agents
Scenario
You're building an agent that observes meetings and triggers actions on key phrases ("schedule a follow-up", "send the contract"). You need transcript chunks in real-time, not after the meeting.
What you get
Stream Meet transcript chunks via WebSocket or use the MCP integration. Connect Claude / GPT directly to live meeting context with sub-second latency.
Quickstart
Get a Google Meet bot live in 5 minutes
The full flow from sign-up to first transcript — no SDK install, no Marketplace approval, no admin consent for unmanaged tenants.
- 1
Get an API key
Sign up at vexa.ai/get-started, or run the self-hosted gateway and mint a key locally.
- 2
Grab the meeting code
From https://meet.google.com/abc-defg-hij take the abc-defg-hij segment as native_meeting_id.
- 3
POST /bots
Send platform=google_meet plus the meeting code. The bot joins within 5-10 seconds.
- 4
Host admits the bot
Most Meet meetings put external joiners in a lobby. The host clicks Admit and the bot becomes active.
- 5
Receive transcripts
Poll GET /transcripts/google_meet/{code} or subscribe to a webhook for push delivery.
How it works under the hood
Google Meet architecture
Vexa joins via headless Chromium in a Playwright-driven bot pod. Audio is captured from the Meet web client, streamed to a Whisper-large-v3-turbo service, and emitted as transcript chunks (typical chunk latency ≈ 1-3 seconds). Identical pipeline in Vexa cloud and self-hosted Kubernetes. No Google API quota, no Workspace OAuth.
Pipeline
Documentation
Go deeper on Google Meet
Reference docs for the Google Meet buyer: URL formats, lifecycle, webhooks, self-host, diarization.
Google Meet platform guide
Supported URL formats, admission lifecycle, common failure modes.
Bots API reference
Full POST /bots payload, status polling, lifecycle hooks.
n8n integration
Wire Meet transcripts to Notion, Slack, Airtable, or any n8n node.
MCP integration
Expose live meeting transcription as a tool for Claude / ChatGPT / any MCP agent.
Webhooks
Push transcripts to your endpoint as chunks land — no polling.
Speaker identification
How diarization works, accuracy bounds, current limits.
Full reference at docs.vexa.ai.
Related reading
Keep going on Google Meet
Tutorials, deep-dives, and adjacent surfaces from across the Vexa substrate.
Google Meet transcription n8n workflow
Auto-transcribe every Meet via n8n + Vexa — HTTP request + Catch Hook patterns.
Claude Desktop MCP setup for Google Meet transcripts
Connect Claude / ChatGPT to live Meet transcription via MCP.
Vexa vs Recall.AI comparison
Apache 2.0, self-host, real-time — how Vexa differs on Google Meet.
Pay only for
what you use
Self-host for free, or let us run it for you. Simple usage-based pricing—no per-seat tax.
Apache 2.0 · Self-host forever
Best for teams who need full control
1 bot · Flat monthly · Everything included
Best for personal use — 1 meeting at a time
No credit card required
Bot infrastructure · +$0.20/hr transcription
Best for teams & API builders · $5 free credit
No credit card required
On-premises, OpenShift, Kubernetes. Dedicated support + SLA.
For self-hosted Vexa bot users. Transcription only — $0.002/min.
Frequently asked questions
What's included in the $0.30/hr Bot Service?
The base rate covers bot infrastructure: audio capture, webhooks, and 12 months of audio storage across Google Meet, Microsoft Teams, and Zoom. Add real-time transcription for +$0.20/hr ($0.50/hr total).
Which plan is right for me?
Choose Individual ($12/mo) if you need one bot for personal use—it includes real-time transcription, storage, and the Dashboard. Choose Pay-as-you-go ($0.30/hr) if you need multiple simultaneous bots or want to pay only for what you use—ideal for teams and API integrations. Both plans include Google Meet and Microsoft Teams support. Self-host for free from GitHub if you need full data control.
How does the free credit work?
Every new account gets $5 in free bot credit—no credit card required. That covers ~16 hours of bot time at $0.30/hr. All features are available: audio capture, transcription, real-time data, and full API access.
Can I self-host Vexa?
Yes. Vexa is Apache 2.0 licensed and fully self-hostable. Deploy with Docker Compose, Kubernetes, or OpenShift on your own infrastructure with complete data sovereignty.
How does Vexa compare to Recall.ai?
Vexa is open source, self-hostable, and up to 40% cheaper. Bot: $0.30/hr vs Recall.ai ~$0.50/hr. Transcription add-on: $0.20/hr vs ~$0.15/hr. See our detailed comparison.
What's the Individual plan for?
The Individual plan ($12/mo) is for single users who need one concurrent bot with unlimited meetings. It includes real-time transcription, recording, storage, REST API, WebSockets, and the UI Dashboard—everything you need for personal meeting intelligence. It supports Google Meet and Microsoft Teams, with Zoom coming soon.
Ready when you are
Start with Google Meet.
First transcript in 5 minutes.
No credit card, no Marketplace approval, no admin consent for unmanaged tenants. Get an API key, POST a meeting URL, receive transcripts.
Apache 2.0 · Self-host or cloud · No credit card required