Video, voice, images and systems meet in one practice. We follow an idea across media until it finds the form that carries it best.
Four directions across moving image, voice, editorial imagery and connected systems. Explore the work as a sequence, each with its own rhythm.
Moving image / 01A moving-image direction where a monumental setting gives a vehicle campaign its scale.
The recording lists this as “Automotive Brand Campaign” under Video · Avatar, dated 2025. The artwork here is a new visual interpretation rather than a frame copied from the original.
Voice / 02A voice-led editorial format, pairing spoken content with a clear written system.
The recording labels this work “Pharma Podcast-Serie” under Audio · Text, dated 2025. We retain the title and medium while creating a separate artwork for this site.
Image / 03A visual language built for both editorial presence and campaign adaptation.
The recording lists “Fashion Editorial” under Bild · Ads, dated 2024. This composition uses form, framing and material contrast to reinterpret that brief.
System / 04A systems perspective on creative operations, where process becomes part of the brand experience.
The recording identifies this as “FinTech Automation” under Workflow · Marketing, dated 2024. No performance claims or client details are inferred from the short clip.
Titles, categories and years are transcribed from the supplied recording. All four artworks are original reinterpretations in this site’s visual language.
The same idea can demand a film, a voice, an image or a connected system. The discipline follows the question.
An independent AI studio working across video, audio, avatars and design. The medium can change; the point of view holds everything together.
Explore the separate sound experiment ↗This interactive cat-sound study comes from a separate MIMO project shared on X. It is presented here as an independent experiment, not as one of the studio cases above.
Choose a short recording. This browser study reads its frequency and volume to animate the sculpture below, directly on your device.
Use a cat sound, a spoken word, or any audio file. Press play and watch the geometry respond. This is an original visualization study inspired by MIMO’s public process, not a reproduction of its full analysis pipeline.
The original MIMO post describes an unusually precise transformation of real cat recordings into interactive geometry.
Cat recordings become 22,050 Hz mono PCM, ready for frame-by-frame analysis.
Meyda extracts RMS energy, spectral centroid, spectral flatness and MFCCs using 2,048-sample Hann windows.
Selected features are smoothed through 84 temporal layers to control density, spread and contour.
Three.js GPU instancing and shaders synchronize voxel movement with the audio playhead.
These are details disclosed by the original project. The interactive visualization on this page uses a lighter Web Audio and Canvas implementation.
The original MIMO X profile publishes the MIMO token contract. Its token is listed on Robinhood Chain, an Ethereum Layer 2. This page can connect an EVM wallet, switch to that network and read your MIMO balance.
Network parameters follow Robinhood’s documentation. The connection here is read-only after wallet access: it never requests a signature, approval or transaction.
Always compare this address with the one in the original project’s X profile.
Project information and network settings are linked to their public sources so the artwork and token identity remain easy to check.
The original post describes the cat-audio pipeline and links to the original interactive experience. The X profile displays the contract address used here.
Read the original post ↗Official network configuration: mainnet chain ID 4663, public RPC, ETH as gas, and the network explorer.
Check network details ↗