Template name:
Script-to-Reel Cinematic Director
https://elevenlabs.io/app/templates/sQr9Wz8QSWXc4x82TkvS
Short description
A reusable Template that turns a structured 4-part script (Hook, Conflict, Value, CTA) plus two reference images into a complete cinematic 30-second reel. Every ElevenCreative model in one pipeline: Gemini LLM, Seedance video, ElevenLabs Voice, Sound Effects, and Music, mixed in the composite for a finished reel or downloaded as separate assets for advanced editing.
Long description / what it does and for whom
Generating a cohesive 30-second cinematic reel usually requires manual, multi-step prompting across disparate tools. Executing this reliably within a single automated pipeline is notoriously difficult.
Script-to-Reel Cinematic Director solves this through a robust 30-node architecture within ElevenLabs Flows. It is built for filmmakers, brand creators, marketers, and educators who want to produce cinematic short-form reels without writing a single prompt for the underlying models.
The Template accepts three inputs: a structured XML-style script with four narrative sections (Hook 5s, Conflict 10s, Value 10s, CTA 5s), an avatar model sheet, and a product model sheet. From there, thirteen LLM nodes read the script, analyze the reference images, and generate optimized prompts for: Seedance 2.0 for video, ElevenLabs Voice for voice-over and narration, ElevenLabs Sound Effects for critical scene-defining sounds, and ElevenLabs Music for the synchronized instrumental score.
The composite node combines all assets with multi-track audio. Novice users get a finished reel they can download and post. Advanced users download the individual assets and assemble them in their preferred NLE (DaVinci Resolve, Premiere, Final Cut).
The Template is fully agnostic. Replace the script and the model sheets, and the same pipeline produces a new reel in any genre, for any brand, any narrative. Thriller, romance, horror, documentary, fantasy, action. The LLMs detect genre and tone from the script's EMOTIONAL_TARGET and SOUND_DESIGN fields, and adapt cinematography, voice delivery, sound design, and music accordingly.
Built for the community to scale high-end production.
Inputs (when defining the Template)
SCRIPT: structured 4-part script in XML format (HOOK, CONFLICT, VALUE, CTA)
AVATAR: character model sheet image (multiple angles, neutral pose, clean background recommended)
PRODUCT: product model sheet image (multiple angles, clean background, key details visible)
Outputs
Final 30-second cinematic reel (composite, ready to download and post)
Individual video clips per section (HOOK, CONFLICT, VALUE, CTA) for advanced editing
Voice-over audio tracks per section (if VO is present in the script)
Sound effects audio per section
Instrumental music score (30 seconds, synchronized)
0
73
Through a fractured found-footage recording, a young woman’s final night in the Amazon reveals what happens when the jungle is watching—and decides you will not leave.
1
34
Ordinary lives unfold in familiar, everyday moments until something shifts. Through the subjective perception of six individuals, reality fractures as they experience epileptic seizures, revealing a deeper truth: the real fear is not the seizure itself, but the constant uncertainty of when it will happen.
By placing the audience inside these moments of loss of control and vulnerability, the film transforms an invisible condition into a visceral, shared experience. In the aftermath, what remains is not fear, but the presence of others who know how to respond. Understanding epilepsy does not stop the condition, but it changes what happens next.
1
65
Built for the Melius Challenge.
Most AI workflows are single-use machines. This one is not. I built a reusable cinematic reel director on @Melius that turns a structured 4-part script (Hook · Conflict · Value · CTA) plus two reference images into a complete 30-second cinematic reel. Any product, any story, any genre.
The example below is YANA by JASSA, a fictional limited edition perfume. Noir adventure in the Peruvian Amazon. Pure visual storytelling. Powered by Claude Opus, Seedance 2.0, and ElevenLabs Music.
Built for the community, not just for me.
#MeliusChallenge #AIFilmmaking #CinematicAI #Seedance #ElevenLabs #AIWorkflow #ShortForm #ContentCreation
1
45