Back to Projects

Generative AI + Media Automation

AI Content Production Pipeline

An end-to-end content workflow that takes an idea from Asana and automatically moves it through script creation, scene generation, visuals, captions, AI voice, FFmpeg rendering, and final storage in Google Drive.

n8nAsanaLLMsAI ImageAI VoiceFFmpegGoogle Drive

Content Pipeline

Idea to video

Generative AI

Live activity

8

Steps

0

Manual

Drive

Output

n8n
Asana
LLMs
AI Image Generation
AI Voice
FFmpeg

Section 01

The Problem

Repetitive production steps

Every short video repeats the same scripting, visual and editing chores.

Too many tools

Ideas, scripts, images, voice and editing all live in different apps.

Slow turnaround

One piece of content can take hours of hands-on work before it ships.

Inconsistent output

Manual assembly makes tone, format and quality drift between posts.

Section 02

The Solution

Content Queue in Asana

Ideas enter the pipeline as simple task cards.

AI Script Generation

A structured script is written from the idea.

Scene Breakdown

The script is split into scenes and transcript beats.

AI Image Generation

Visuals are generated for every scene.

Caption Creation

Captions and copy are produced alongside the visuals.

AI Voiceover

Narration is synthesised in a consistent voice.

Automated Assembly

FFmpeg stitches visuals, voice and captions together.

Drive Storage

The finished asset lands in Google Drive, ready to post.

Section 03

How It Works

Asana IdeaStep 1
AI ScriptStep 2
Scene BreakdownStep 3
Image GenerationStep 4
Caption GenerationStep 5
AI VoiceStep 6
FFmpeg RenderingStep 7
Google DriveStep 8

Section 04

System Architecture

Asana

Content queue and trigger

n8n

Runs every stage in order

LLM

Scripts, scenes and captions

Image AI

Scene visuals generated on demand

Voice AI

Narration audio

FFmpeg

Rendering and assembly

Google Drive

Final storage and delivery

Generative layer

Language, image and voice models sit inside one orchestration — the same idea flows through each of them without a human moving files between tools.

ScriptVisualsVoice

Section 05

Tech Stack

n8n

Pipeline orchestration

Asana

Content queue

LLMs

Scripts and captions

AI Image Generation

Scene visuals

AI Voice

Narration audio

FFmpeg

Video rendering

Google Drive

Asset storage

Section 06

What This Project Demonstrates

Generative AI

Text, image and voice models used together.

Multi-Step Automation

Long chains that run unattended end to end.

Media Automation

Programmatic rendering and asset delivery.

Workflow Orchestration

Reliable sequencing across many services.

Let's build

Want something like this for your business?

I build custom AI and automation systems around real business workflows.