V-Transcoder: Video Processing Microservices Platform

A Python microservices application for video processing, including conversion to MP3, speech-to-text transcription, and openai translation services.

Prerequisites

• Docker
• kubectl
• minikube
• PostgreSQL (running on localhost:5432)
• MongoDB (running on localhost:27017)

Create .yaml files for the required secrets:

auth-secret → contains POSTGRES_PASSWORD, JWT_SECRET

notification-secret → contains GMAIL_ADDRESS, GMAIL_PASSWORD

openai-secret → contains OPENAI_API_KEY

Example structure:

apiVersion: v1
kind: Secret
metadata:
name: auth-secret
type: Opaque
stringData:
POSTGRES_PASSWORD: 'your-db-password'
JWT_SECRET: 'your-jwt-secret'

Quick Start

1. Start minikube:

   make start-minikube

2. Build images:

   make build-all

3. Deploy services:

   make deploy-all

4. Start tunnel (in another terminal):

   make tunnel

5. Add to /etc/hosts:

   127.0.0.1 mp3converter.com
   

6. Access: http://mp3converter.com

Services

• Auth: User authentication and authorization (port 5000)
• Gateway: Main API gateway and request routing (port 8080)
• Converter: Video to MP3 conversion service
• Transcriber: Speech-to-text transcription using Whisper AI
• Translator: AI-powered text translation using OpenAI GPT
• Notification: Email notification service
• RabbitMQ: Message queue for service communication