Deployment Guide

Learn how to deploy WhispWizard in production environments

Frontend (Vercel)

Build Command

npm run build

Output Directory

/client/dist

Environment Variable

VITE_API_BASE_URL=https://yourbackend.render.com

Backend (Render/AWS)

Start Command

uvicorn main:app --host 0.0.0.0 --port 8080

Recommended Instance

AWS GPU instance (g4dn.xlarge)

Environment Variables

WHISPER_MODEL=medium
MAX_FILE_SIZE=100MB
API_KEY=your-secret-key

Important Notes

  • For production use, always enable HTTPS and secure your API endpoints with authentication
  • The Whisper model requires significant memory (medium model needs ~5GB RAM)
  • For high-traffic applications, consider implementing rate limiting and request queuing
  • Monitor your GPU utilization and scale accordingly during peak loads