When AI can generate a finished track in seconds, what matters most about music is the experience of making it, from discovering a sound to learning to play it and sharing it with other people. Xionia Labs researches artificial intelligence that deepens that experience, rather than replacing it.
Research in progressA research programme, not a product announcement.
The question
Generated music is going to be everywhere, in advertising, in background listening and in soundtracks made for a single listener. We don’t believe that ends the human desire to make music.
People still learn the piano, paint and write poetry long after technology offered easier ways to get the result, because creating something and consuming it are different experiences. So our research starts from one question. How can creating, playing and sharing music become more rewarding for the person doing it?
“Audio plug-ins are becoming less about music production and more about enabling human creativity, fun, connection and expression.”
The idea
Imagine describing the instrument you want to play.
“I want something that feels like a cross between a cello and an evolving analog synthesizer. I want it to respond to the pressure of my fingers, become unstable when I play aggressively, and have a beautiful, haunting quality when played gently.”
Working with a deep synthesis engine, it designs the oscillators, filters, resonators, envelopes and modulation, and the expressive relationships between them. You don’t need to understand any of it to begin playing.
What you have is a real instrument, with characteristics to discover, techniques to learn and expressive possibilities to master.
Research areas
Each area starts from a question about the person making the music, not the music produced.
Expressive control
Oscillators, filters, envelopes and modulation stay under the hood. On top, we’re researching AI that builds a few expressive, high-level controls, such as tension, fragility, energy, movement and intimacy, so a sound becomes an instrument that answers the way you play. Experienced musicians can still reach everything underneath.
Transparent assistance
Ask for a more unstable, vintage character, and an assistant should say exactly what it did, from the drift it added and the phases it moved to the saturation it introduced. Every change stays open to keep, adjust, reject or automate. Assistance should be optional, visible and reversible by design.
Expression and gesture
Pressure, pitch movement, phrasing and gesture carry musical meaning. We’re researching instruments that respond to them organically, so a performance shapes the sound as it unfolds.
It extends to experimental inputs. We’re exploring EEG integration, so that signals from compatible headsets could be mapped to musical controls, with the connection visible, adjustable and optional like any other.
Mastery and discovery
The instruments people love respond consistently enough to build real skill, and richly enough to keep revealing something new. We’re studying how to design for both, and how to tell whether an instrument is rewarding to play, not just how it sounds.
Playing together
Some of music’s best moments happen between people. We’re exploring shared instruments that friends can play together over the internet, where the fun comes less from finishing a track than from what happens when people play off one another.
Each person plays a different part of the same instrument, one the notes, another its timbre, another its rhythm and another its character. No single player controls it completely.
Everyone brings an instrument of their own making, and the instruments listen to each other. One player’s harmony reshapes another’s sound; another’s rhythm sets the first in motion. Each new player changes the ensemble.
Groups explore musical environments that come alive as they play, with challenges, discoveries and the occasional moment of glorious chaos. Here the research meets the interactive experiences of Xionia Studios.
Someone who has never played an instrument can join in with a few expressive controls or a touchscreen, while an experienced musician dives into the synthesis underneath. AI takes care of the technical side, such as harmonic compatibility and the connections between instruments, but it never plays the music for anyone.
Instruments created together can be saved, shared and developed further, so what a group discovers in one session becomes material for the next.
Latency across the internet, keeping players in time and making instruments answer each other musically are hard problems to solve. That’s why this is research.
Measures
Speed and polish are easy to measure. We hold this research to the five questions Xionia Audio asks of every instrument.
Creativity
Does it lead you to ideas you wouldn’t have found on your own?
Expression
Can you put your feelings and personality into the sound?
Enjoyment
Is it satisfying to explore, to play and to master?
Connection
Can you share what you make, and make music with other people?
Growth
Does it grow with you, and help you find a style of your own?
Our goal is more people making music.
Research vision
Stephen Ankelen, Founder and CEO, Xionia
“Technology will change profoundly in the coming years. The software we use, the interfaces we interact with, and even the ways we access digital experiences may become almost unrecognisable.
But the reasons people create music will remain remarkably familiar.
We will still want to explore sounds, express emotions, develop skills, discover new possibilities and share meaningful experiences with others.
As artificial intelligence makes creating music increasingly accessible, we believe the opportunity for audio technology extends far beyond helping people produce better results.
It is about creating more inspiring, expressive and enjoyable ways to experience music-making itself.
We envision intelligent instruments that people can invent, play and master. Creative environments that adapt to individuals rather than requiring individuals to adapt to software. And connected musical worlds where friends and musicians can discover, improvise and create together, wherever they are.
Our research explores how powerful audio technologies and artificial intelligence can come together to empower these experiences, making the technology less visible and human creativity more central.
Because the future of music isn’t simply about what machines can create for us. It’s about what they can empower us to create, discover and experience together.”
From research to instruments
Research earns its place in a product when it makes music-making more rewarding. Then it reaches musicians through the instruments of Xionia Audio, and players through the interactive experiences of Xionia Studios.
One example is in development now. Optional AI tools for StarShaper, Xionia Audio’s instrument creator, are built to help you start, refine and map an instrument, with every suggestion open to inspection and change. The technology behind them is the subject of our second programme, Xionia Intelligence.
This page describes research, not products. Anything that becomes a product will be announced by the Xionia division that makes it, with what it does and when it’s available.
We’d like to hear from musicians, music educators, researchers and developers who care about the future of making music.
