How to Self-Host Your Own AI Meeting Assistant

Your meeting recordings contain sensitive information—project plans, financial discussions, HR conversations. Why send that data to third-party servers? Self-hosting gives you complete control.

This guide shows you how to run TellMeMo, a full-featured AI meeting assistant, on your own infrastructure. Setup takes about 10 minutes.

Why Self-Host Your Meeting Assistant?

Privacy & Security

With cloud services like Otter.ai or Fireflies, your meeting recordings are uploaded to their servers. They promise security, but you're trusting them with:

• Confidential business discussions
• Employee performance reviews
• Financial planning conversations
• Legal and compliance matters
• Product roadmap discussions

Self-hosting means your data never leaves your infrastructure. Period.

Compliance

Self-hosting can help meet regulatory requirements:

• HIPAA — Healthcare organizations can keep PHI on controlled systems
• GDPR — EU data stays in your EU datacenter
• SOC 2 — You control the entire audit trail
• Industry regulations — Finance, legal, government requirements

Cost Control

Commercial tools charge $15-30 per user per month. Self-hosting TellMeMo costs:

• Software: Free (open source)
• AI APIs: ~$0.01-0.05 per meeting
• Server: Use existing infrastructure or ~$20/month VPS

For a 20-person team, that's $400/month with Otter vs ~$30/month self-hosted.

System Requirements

Step-by-Step Installation

Step 1: Install Docker

If you don't have Docker installed:

# Ubuntu/Debian
curl -fsSL https://get.docker.com | sh
sudo usermod -aG docker $USER

# macOS
# Download Docker Desktop from https://docker.com

# Verify installation
docker --version
docker compose version

Step 2: Clone the Repository

git clone https://github.com/Tell-Me-Mo/tellmemo-app.git
cd tellmemo-app

Step 3: Get Your API Keys

TellMeMo needs API keys for AI services. The software is free—you pay only for what you use.

Required: Anthropic API Key

1. Go to console.anthropic.com
2. Create an account and add billing
3. Generate an API key

Optional: Hugging Face Token

1. Go to huggingface.co/settings/tokens
2. Create a token with "read" permissions
3. Used for downloading embedding models

Step 4: Configure Environment

Create your configuration file:

# Create .env file
cat > .env << 'EOF'
# Required
ANTHROPIC_API_KEY=YOUR_ANTHROPIC_API_KEY_HERE
HF_TOKEN=YOUR_HUGGINGFACE_TOKEN_HERE

# Optional: Generate a secure JWT secret for production
# JWT_SECRET=$(openssl rand -hex 32)
EOF

Step 5: Start the Services

# Start all services
docker compose up -d

# Watch the logs (first startup takes 3-5 minutes)
docker compose logs -f

The first startup downloads AI models and initializes databases. This takes 3-5 minutes. Subsequent starts are much faster.

Step 6: Access the Application

Open your browser and go to:

Create an account and start uploading meetings!

Architecture Overview

TellMeMo runs several services in Docker containers:

Production Deployment

For production environments, consider these additional steps:

1. Use HTTPS

Set up a reverse proxy (nginx, Traefik, or Caddy) with SSL certificates:

# Example with Caddy (automatic HTTPS)
meetings.yourcompany.com {
    reverse_proxy localhost:8100
}

api.meetings.yourcompany.com {
    reverse_proxy localhost:8000
}

2. Secure Your Database

# In .env, change default passwords
POSTGRES_PASSWORD=your-secure-password-here
JWT_SECRET=your-32-byte-hex-secret

3. Set Up Backups

# Backup PostgreSQL daily
docker exec tellmemo-postgres pg_dump -U pm_master pm_master_db > backup.sql

# Backup Qdrant vectors
docker exec tellmemo-qdrant qdrant-backup /backups/

4. Monitor Resources

Use Docker's built-in monitoring or tools like Prometheus/Grafana:

# Check container health
docker compose ps
docker stats

Troubleshooting

Containers Won't Start

# Check logs for errors
docker compose logs backend

# Restart all services
docker compose down
docker compose up -d

Database Connection Errors

# Wait for database to be ready
docker compose logs postgres

# Reset database if needed
docker compose down -v  # Warning: deletes all data
docker compose up -d

API Key Errors

# Verify API key is set
cat .env | grep ANTHROPIC_API_KEY

# Restart backend after changing .env
docker compose restart backend

Frequently Asked Questions

Next Steps

Now that your self-hosted meeting assistant is running:

1. Upload your first meeting — Try a transcript or audio file
2. Generate a summary — See AI-powered meeting intelligence
3. Try RAG search — Ask questions about your meetings
4. Enable real-time mode — Use live Q&A in your next meeting
5. Invite your team — Add users to your organization

Related Articles

• Best Open Source Meeting Transcription Software 2026
• Otter.ai Alternatives: Open Source & Privacy-Focused Options
• Real-Time Meeting Intelligence: How It Works