Yutaka Makino

WORKS

Interstice	Latent Space	Variations	Confabulation	System of Transfer
Canon for Computational Utterance	Continuum	Technical Ensemble	Formation	V
The Program II	Displacement	Foldings	Void II	The Program
Contingent Matter	Structure of Behaviors	Linear Decomposition	Radiant	Void
RGB	Gradient	Waveguide	Atmosphere	The Conditions of the Process
Temporal Object #1	Temporal Object #2	Systema for WFS	Pulsar	Conflux
Structural Variance	Amorphous for WFS	The Blackest Flux

Interstice
2026

Installation

Dimension Varies

Multichannel Surround System

Interstice is a multichannel sound installation composed of vocal residues generated through AI timbre transfer, a machine learning process that transforms the sonic characteristics of one sound into another. Unresolved by the transformation process, these sounds occupy an indeterminate region between noise and voice. Distributed intermittently throughout the space, their unstable vocal quality invites a sustained mode of listening that gradually extends toward the shared acoustic space. Interstice brings the listener’s own breathing, the presence of others, and the site’s ongoing acoustic activity into the same field of attention.