Vexa: Self-Hosted Meeting Intelligence Platform with Real-Time Transcripts

Introduction

Vexa is an open-source, self-hostable meeting intelligence platform that allows you to deploy bots into virtual meetings (Google Meet, Microsoft Teams, and soon Zoom) for real-time transcription and translation. Built for data sovereignty, Vexa offers full control over your data, making it ideal for regulated industries and privacy-conscious teams. Its modular architecture and robust API enable seamless integration and scaling from edge devices to millions of users.

Installation

Vexa offers flexible deployment options, from a fully hosted service to complete self-hosting.

Option 1: Hosted (Fastest)

You can start immediately by grabbing your API key from the official Vexa dashboard.

Option 2: Self-host with Lite Container (Single Container, No GPU)

This compact version of Vexa depends on an external database and transcription service.

docker run -d \
  --name vexa \
  -p 8056:8056 \
  -e DATABASE_URL="postgresql://user:pass@host/vexa" \
  -e ADMIN_API_TOKEN="your-admin-token" \
  -e TRANSCRIBER_URL="https://transcription.service" \
  -e TRANSCRIBER_API_KEY="transcriber-token" \
  vexaai/vexa-lite:latest

For complete setup examples and more configurations, refer to the official documentation.

Option 3: Self-host with Docker Compose

Ideal for development, this option allows you to run Vexa with Docker Compose.

git clone https://github.com/Vexa-ai/vexa.git
cd vexa
make all            # CPU by default (Whisper tiny) — good for development
# For GPU:
# make all TARGET=gpu    # (Whisper medium) — recommended for production quality

A full guide is available in the project's docs/deployment.md.

Examples

Vexa's API allows you to easily send bots to meetings and retrieve transcripts.

Send a bot to Microsoft Teams

curl -X POST https://<API_HOST>/bots \
  -H "Content-Type: application/json" \
  -H "X-API-Key: <API_KEY>" \
  -d '{
    "platform": "teams",
    "native_meeting_id": "<NUMERIC_MEETING_ID>",
    "passcode": "<MEETING_PASSCODE>"
  }'

Send a bot to Google Meet

curl -X POST https://<API_HOST>/bots \
  -H "Content-Type: application/json" \
  -H "X-API-Key: <API_KEY>" \
  -d '{
    "platform": "google_meet",
    "native_meeting_id": "<MEET_CODE_XXX-XXXX-XXX>"
  }'

Get transcripts over REST

curl -H "X-API-Key: <API_KEY>" \
  "https://<API_HOST>/transcripts/<platform>/<native_meeting_id>"

For real-time streaming with sub-second delivery, a WebSocket API is also available.

Why Use Vexa?

Data Sovereignty: Vexa is built for data sovereignty, allowing full self-hosting of the platform, database, and transcription services on your own infrastructure. This is crucial for regulated industries and organizations with strict privacy requirements.
Real-time Transcription & Translation: Leverage real-time multilingual transcription supporting 100 languages with Whisper, along with instant translation across all supported languages.
Flexible Deployment: Choose from fully hosted, GPU-free self-hosting (using external transcription services), or full self-hosting to perfectly match your privacy and infrastructure needs.
Build Custom Solutions: The Vexa API provides powerful abstractions, enabling developers to build sophisticated meeting assistants, custom integrations, and applications rapidly, similar to Otter.ai or Fireflies.ai.
Multi-Platform Support: Seamlessly integrate with Google Meet and Microsoft Teams, with Zoom support coming soon.