Multiscene Cinematic Video Factory
55 node cinematic ad pipeline with Seedance, Flux and voiceover in one flow
Turns a short brief into three cinematic scenes with voiceover, music and captions. Built for agencies that need ad ready video without a production crew.
A system that works while you are not looking
This system turns a single Telegram message into a complete multisene advertisement. You send a short idea like a product and a mood, and you get back three coordinated cinematic scenes with matching voiceover, background music and burned in captions. It was designed for agencies and brands that need launch ready video at the speed of chat. The final output is a publish ready file plus a schedule entry, so the same idea can be reused for multiple campaigns without rebuilding anything.
A launch ready ad in minutes instead of days, with every asset tracked in Sheets for reuse.
Outcome
A Telegram message starts the flow. The text is split and sent to an AI agent that acts as a prompt engineer and voiceover writer. The agent creates three distinct scene prompts and a short script, keeping style and character consistent across all scenes. Those prompts are sent in parallel to Replicate with the ByteDance Seedance model. Each scene is polled until the video is ready, then collected. A separate branch generates the voiceover with ElevenLabs and prepares the background music. The three scenes, voiceover and music are merged, captions are added through JSON2Video, and the result is logged in Google Sheets as a bulk scheduler. The finished video and schedule link are sent back to Telegram for review.
Trigger
Inbound signal starts the flow
Orchestrate
Models, tools and branches run
Deliver
Result logged and returned
Who it is for
Creators, media teams and brands that repurpose one asset into many at scale.
What you save
Hours become minutes per task, handoffs stay consistent in Sheets or CRM, and the workflow file is yours to own. Import, add your keys and run.