STORIES IN MOTION
In developmentAI video production
CUTROOM
Gather a story. Bring it to the screen.
A local short-form production studio that turns sourced stories into dialogue, voices, subtitles, and scenes.
HOW IT COMES TOGETHER
The core workflow- 01
Stories and sources
- 02
Dialogue and scenes
- 03
Visuals and voices
- 04
Video rendering
01 / THE QUESTION
The problem we set out to solve.
A story can lose its sources and distinctive voices as it becomes a video. Creating dialogue, scenes, and audio separately makes every revision a timing exercise.
02 / OUR APPROACH
Our approach.
We collect stories with their sources and develop dialogue and scenes together. A production workspace lets creators adjust images, motion, audio, and subtitles before exporting an MP4.
- 01
Stories with sources
Original posts and supporting material remain connected to the script so creators can review them while editing.
- 02
Character voices and subtitles
Multi-character TTS, dialogue captions, and timing controls give each voice its place in the scene.
- 03
Layer-based scene editing
Images and photographs are arranged with layer motion and an adjustable audio mix.
- 04
Local rendering
Job cancellation and resumption, asset deduplication, MP4 export, and 4K composition are handled with FFmpeg.
03 / INTELLIGENCE AT WORK
Where AI does the work.
LLM-assisted scripting, image generation, and character TTS are connected to the production process. Source material and AI-generated assets remain distinguishable, while creators edit the scenes and language.
04 / WHERE WE ARE
Where the project stands.
A local tool with a React workspace, Python rendering backend, and macOS launcher. External AI features depend on the services configured for the project.
