You move your jaw a fraction of a millimeter. You do not speak. Nobody hears a sound. But something did: a sensor the size of a grain of rice, tucked in the arm of a pair of glasses, just felt the faint magnetic field your facial muscle threw off when it fired. No wire touched your skin. The signal was always there - your body has been broadcasting it your whole life. Until recently, nothing small enough could catch it.
That grain of rice is Sonera's whole thesis: the most useful data about a human being is magnetic, and it has been hiding in plain sight because the only instruments that could read it needed liquid helium and a shielded room.
Every time a muscle contracts, electrical currents run through it. Those currents produce a magnetic field. Electrode-based sensors - the EMG and EEG hardware neurotech has leaned on for decades - measure the electricity, which means they need firm skin contact, conductive gel, and a tolerance for everything that goes wrong when skin gets sweaty or hairy or simply moves.
Sonera measures the magnetic field instead. Its S1 chip performs magnetomyography: reading the magnetic signature of muscle and nerve activity from a small distance, no contact required. The upshot is a sensor that is far less fussy about the surface it sits near - and small enough to disappear into an earbud, a necklace, a pair of glasses, or a smartwatch.
The company's stated goal is almost stubbornly plain: make neural and neuromuscular data as ordinary to capture as a heart rate. Not a lab result. A background reading your devices take without you thinking about it.
That framing matters. Sonera is not selling a headset or a wristband. It is selling the sensor underneath - the missing piece that makes those products possible at consumer scale.
We are trying to detect brain activity using cheaper, faster methods that are still high-performance.
Muscle and nerve activity generates tiny magnetic fields - orders of magnitude fainter than the Earth's, far too weak for ordinary sensors.
Sonera uses acoustically driven ferromagnetic resonance: high-frequency sound waves interacting with magnetic thin films to detect those faint fields - at room temperature.
The S1 turns that magnetic signature into a clean, high-fidelity signal a device can decode into speech, gestures, or diagnostics.
Figures per IEEE Spectrum reporting on conventional magnetoencephalography. Sonera's cost target is directional, not a published price.
The flagship: a tiny magnetic field sensor doing non-invasive muscle sensing without skin contact. The core component everything else is built on.
Magnetometers in a headphone decoded unspoken words at up to 86% per-user accuracy; a necklace form factor handled both silent and overt speech in ambient conditions.
Facial-muscle and neuromotor signals become control schemes for glasses, earbuds, and immersive interfaces - input without a controller.
Room-temperature, shield-free magnetometry aimed at affordable brain imaging - and tracking neuromuscular biomarkers for conditions like neurodegeneration.
Met her co-founder as an electrical engineering PhD student at UC Berkeley. Leads the company's push to make neural data broadly accessible.
Berkeley PhD in magnetics and semiconductor devices; the technical architect of Sonera's sensing approach.
Labanowski's doctoral advisor. Brought the deep device-physics pedigree that made the sensor plausible in the first place.
They had the clarity to acknowledge past approaches would not work.
Sonera Magnetics founded in Berkeley, commercializing years of UC Berkeley research in magnetics and semiconductor devices.
Exits stealth with an $11M seed round led by Amplify Partners, with Material Impact, Spark Capital, Boom Capital, Abstract Ventures, Two Bear Capital and others.
Launches "The Magnetic Moment," a public talk series on magnetics and neurotechnology.
Publishes research on silent speech recognition with wearable magnetometers - 86% per-user accuracy, plus a necklace that decodes speech in real-world conditions.
Back to those glasses, and that jaw moving a fraction of a millimeter. Before Sonera, the honest answer to "can a device read that without touching you?" was no - not at a price or a size anyone would put on their face. Reading biomagnetics meant a machine cooled colder than deep space, wrapped in metal shielding, priced like a house.
Sonera's wager is that the machine was never the point. The signal was. And once you can catch the signal with a chip that runs warm and costs little, the silent word you didn't say out loud becomes a command, the clenched muscle becomes a click, and the early tremor of a neuromuscular disease becomes something a wearable might notice before you do.
That is still a wager, not a finished product line. Sonera is a small team turning a lab result into a shippable part, and the distance between "86% in a study" and "in ten million earbuds" is exactly the distance every hardware company has to walk. But the scene has already changed. The magnetic whisper your body has always made is, for the first time, something a small machine can hear - and quietly answer.