I design interactive systems that translate hidden dynamics into sound, working across electroacoustics, psychoacoustics, and real-time computation. My work holds together two crafts that usually sit apart, the sound design that decides how something should feel to hear and the engineering that makes it run in real time.
An audio engine that gives electric cars a voice by reading real telemetry data including RPM, load, and speed via OBD2. It translates these metrics in real time into a synthesized engine sound, utilizing a custom hybrid of wavetable and granular synthesis built in Max MSP and rigorously field-tested on the road.
Turns Binance's order flow into a continuous, low-fatigue sound field: a "sixth sense" for the market, to listen to in the background instead of staring at a chart. Five metrics drive three synthesis engines and a reverb network, all in real time. Built with RNBO and SvelteKit.
Electric cars are silent, but this engine gives them an intentional and informative voice synthesized in real time from the car's own telemetry.
Electric vehicles removed the engine note that drivers unconsciously rely on to sense speed and effort. The project rebuilds that channel on purpose: it reads live vehicle telemetry over OBD-II and turns it, in real time, into a synthesized engine sound designed to inform without fatiguing. The system has been deployed and validated on the road.
A Python bridge polls the ECU and streams RPM, engine load, and speed as OSC messages at 20 Hz to a cross-platform Max MSP engine tuned for low latency.
Two synthesis layers run in parallel and are then fused into a single, believable acoustic object.
Wavetable: the melodic core. A dual-engine wavetable carries the harmonic identity: a stable major-third dyad at low speed that opens to a perfect fifth as you accelerate, with a richer seventh-chord layer that shifts from major to minor past 100 km/h, tension that climbs with velocity.
Granular: the organic texture. Thirty-six just-in-time grain voices read from a 22-second buffer mapped one second per 10 km/h: warm, harmonic grains at low speed; dense, rough, mechanical grains at high speed.
FX matrix: the glue. Waveshaping, a resonant filterbank that simulates the cavity of a physical chassis, and an early-reflection network place both layers in one shared physical space.
Every continuous input is ramp-smoothed to eliminate zipper noise, with the granular buffer capped at 22 seconds and held entirely in RAM, while grain voices compute only while a grain is actually sounding. The algorithm was validated across 18 recorded driving scenarios including idle, cruise, acceleration, uphill load, and highway before being tested live on the road.
A continuous, low-fatigue auditory display that turns live Binance order flow into a "sixth sense" for the market.
SoniChain converts a live Binance stream into sound you can keep in the background of attention for hours without fatigue. The goal isn't to encode data as beeps, but to design an ergonomic auditory display: every mapping is chosen against a psychoacoustic rationale rather than a naive data-to-pitch rule. The DSP is authored in RNBO, exported to WebAssembly, and hosted in a SvelteKit app that also ships as a native desktop build via Tauri.
Metrics are computed off the main thread in a TypeScript Web Worker and pushed to the DSP through a strictly one-way control path with no feedback loop from the audio back into the data layer.
This is the heart of the design, where market semantics become perceptual attributes.
Three switchable engines give the display its timbre: a glassarmonica (inharmonic subtractive resonance), an FM pad (a warm, stable four-operator bell) and a bowed string (a Karplus-Strong waveguide). Each runs eight-voice polyphony, so price transients trigger discrete, overlapping notes to create a harp-like decay rather than a siren glissando. A Schroeder-Moorer reverb supplies the shared space, its tail length tracking volatility.
Intensity becomes amplitude, imbalance becomes physical space, turbulence becomes acoustic persistence, a monitoring instrument that doesn't fatigue.
Prices move in musical intervals, never glissando; sell and buy pressure are split hard left and right to keep the stereo center clear; denser flow sounds closer and brighter; volatile markets leave a reverberant wake. An adaptive scaler re-ranges the display as regimes change, and a logarithmic volume curve lets you dial it from a discreet macro-event alarm all the way to full micro-structure immersion.
I build systems that let people hear data instead of watching it. My work sits where electroacoustics, psychoacoustics, and real-time computation meet, turning signals that are normally invisible into sound a listener can hold at the edge of awareness.
I'm a composer and audio developer finishing a Master's in Electronic Music at the Conservatorio Tartini in Trieste, where I also took my Bachelor's, on a research track built around interactive and data-driven audio. My practice runs between two poles. One is the engineering needed to capture and process live signals at low latency; the other is the sound design needed to make the result musical, legible, and comfortable to live with. Every mapping decision begins as a perceptual question, what an experience should feel like to hear, and only then becomes a technical one.
That instinct comes from years as a multi-instrumentalist, trained across percussion, saxophone, guitar, and piano. Playing rooted my ear in the physical behaviour of sound long before I approached it through code, and it still shapes how I judge what I make. Alongside my research I keep an active compositional practice in acousmatic and electroacoustic music, and I treat critical listening as its own discipline rather than a final polish.
My interests fall under sonification and auditory display, the corner of Sound and Music Computing concerned with making information audible without making it intrusive. My next step is graduate study in that field, and work where a composer's ear drives the aesthetic and perceptual side of applied audio.
Code on GitHub, sound on YouTube — pick a signal path.