Back to Portfolio
Automation
n8n
Generative AI
Media Pipeline
AI Video Automation Pipeline
An end-to-end n8n pipeline for generating, assembling, publishing, and tracking YouTube videos and Shorts across multiple languages.
About the Project
We designed and implemented a fully automated content pipeline for creating and publishing YouTube videos and Shorts around a chosen theme.
The workflow orchestrated idea generation with ChatGPT, image creation with Google's Nano Banana model, voice generation with ElevenLabs, music generation with Suno, and video assembly with ffmpeg, all coordinated through n8n.
We also built persistence and tracking layers so the system could record what had already been generated and published, avoid repetition, and support a translation layer capable of producing the same content in additional languages.
Key Features
- n8n orchestration for end-to-end video and Shorts production
- ChatGPT-powered idea generation and scripting around a defined content theme
- Image generation workflow using Nano Banana with a defined visual style guide
- Voice generation with curated ElevenLabs voice selection and cadence tuning
- Music generation with Suno guided by consistent style constraints
- ffmpeg-based media assembly and publishing-ready video output
- Publication tracking to avoid duplicate topics and preserve content history
- Translation layer for multilingual video generation
Technologies Used
n8n
ChatGPT
Nano Banana
ElevenLabs
Suno
ffmpeg
Results
Automated
Video ideation-to-publishing workflow
Consistent
Thumbnails, voice, and music style across outputs
Higher
CTR potential through title and thumbnail optimization
Lower
Audience churn through better hooks and pacing