Joshua Oluseyi

Image to Video Studio

Media Superagent — Offline Production

All browser functions run locally after page load: image-to-video motion, titles, image processing, soundtrack selection, video capture, and conversational commands. Neural diffusion and true model-based text-to-video require an optional GPU worker.

Conversational media controller

Ready for offline editing.

Supported: duration, motion, zoom, titles, brightness, contrast, saturation, audio volume, preview, export, print, and reset. Unknown actions are rejected, never interpreted as arbitrary code.

Image to video / photo animation

Animates the supplied image with camera movement (Ken Burns effect). This is not neural generation of new objects or scenes.

Physical output & connected devices

Print a processed image or save it for compatible printers, cutting/engraving applications, projectors, video screens and authorized media devices.

Device deployment or fabrication requires specific drivers, material and human authorization. This studio does not claim to convert digital media into physical objects by itself.

Neural engines (optional)

GPU host integrations: Stable Video Diffusion image-to-video, local text-to-image Diffusers, Ollama language commands, existing OpenVoice/RVC voice processing. Model installation is separate and depends on license, disk and GPU.

See the local-ai/superagent folder inside the plugin ZIP. No paid API key is necessary for models installed locally.

Share this page by QROpen link