Hathcock Homelab Wiki

Podcast-Studio

2 min read

Podcast Studio (Video Recording Platform)

← Back to Index

IP: 192.168.1.XXX | Repo: https://github.com/jhathcock-sys/podcast-studio

Overview

Self-hosted video podcast recording platform for D&D sessions and podcasts. Supports 4K recording, multi-track audio, remote guests, and live streaming.

Access URLs

ServiceURLPurpose
Frontendhttp://192.168.1.XXX:8080Web application
APIhttp://192.168.1.XXX:3333Backend API
LiveKitws://192.168.1.XXX:7880WebRTC server
MinIO Consolehttp://192.168.1.XXX:9001Storage management

Tech Stack

  • WebRTC Server: LiveKit (real-time communication)
  • Frontend: React 18 + TypeScript + Vite
  • Backend: Node.js/Express
  • Storage: MinIO (S3-compatible)
  • Processing: FFmpeg (post-processing)
  • TURN Server: Coturn (NAT traversal)

Repository

Architecture

Hybrid LiveKit + Double-Ended Recording:

  • LiveKit handles real-time WebRTC and RTMP streaming
  • Local recording on each browser for true 4K quality
  • MinIO stores recordings
  • FFmpeg post-processes with sync and normalization

Storage Planning

QualityPer Hour6-Person 1hr Session
1080p~3.6 GB~21.6 GB
4K~11.25 GB~67.5 GB

Recommendation: Plan ~70 GB per 1-hour 6-person 4K session

Retention Policy

  • Raw uploads: 7 days (auto-delete after processing)
  • Processed files: 90 days in MinIO
  • Archive: Move to NAS-Synology for long-term storage

Network Configuration

Required Firewall Rules

# LiveKit
sudo ufw allow 7880/tcp comment "LiveKit API"
sudo ufw allow 7881/tcp comment "WebRTC TCP"
sudo ufw allow 7882/udp comment "WebRTC UDP"
 
# TURN Server
sudo ufw allow 3478 comment "TURN"
sudo ufw allow 10000:20000/udp comment "TURN relay"
 
# Application (local network only)
sudo ufw allow from 192.168.1.XXX/24 to any port 8080 comment "Podcast Frontend"
sudo ufw allow from 192.168.1.XXX/24 to any port 3333 comment "Podcast API"
sudo ufw allow from 192.168.1.XXX/24 to any port 9000:9001 comment "MinIO"

Development Phases

  • Phase 1: MVP - Basic 2-person video call with recording
  • Phase 2: Multi-track recording (up to 6 participants)
  • Phase 3: Post-processing pipeline (sync, normalize, multi-track export)
  • Phase 4: Live streaming (YouTube/Twitch)
  • Phase 5: Scene switching & polish (layouts, overlays, waiting room)

Status

Created: 2026-02-03 Status: Initial setup complete, ready for deployment

TODO

  • Copy .env.example to .env and configure secrets
  • Generate LiveKit API credentials
  • Apply UFW firewall rules
  • Deploy stack with docker compose up -d
  • Test basic 2-person video call
  • Configure Nginx Proxy Manager for SSL (podcast.yourdomain.com)

Graph View

Backlinks