My Electron and Web Audio terminal experiment that turns command-line activity into an immersive generative audio experience.
I built echoSH as a synesthetic terminal experiment: a desktop command-line interface where keystrokes, commands, errors, and process activity generate sound in real time. It explores what a developer tool feels like when terminal feedback becomes auditory as well as visual.
echoSH is a desktop terminal experiment that translates command-line interaction into generated audio. Every keystroke, command, and process event can become part of a soundscape, turning the terminal from a silent text surface into an instrumented feedback loop.
Terminals are information-dense but almost entirely visual. They tell you what happened after you look back at the screen. I wanted to explore whether subtle, real-time audio feedback could make command-line work feel more immediate, expressive, and legible without replacing the normal terminal workflow.
I worked on the concept, product direction, Electron application structure, terminal interaction model, and generative audio behavior. The project is both a developer tool and an experiment in how software can use sensory feedback without becoming distracting.
echoSH is built with TypeScript, Electron, React, Vite, Node.js, and the Web Audio API. The project uses generated sound rather than bundled audio samples, which keeps the sound system programmable and tied to runtime events.
The Electron shell provides the desktop application boundary. The React interface handles the terminal surface and control UI. Command and input events are interpreted into semantic cues, which are then mapped to Web Audio synthesis patterns for keystrokes, successful commands, errors, and diagnostic feedback.
Experimental and early-stage. The public repository documents the core concept, feature direction, local development flow, and roadmap for richer synthesis, sound blueprints, and installable releases.
Repository and related topic hub: Full-stack development.
