Give every device the sense of hearing

Sonaid is an INRIA spin-off building the embedded Sound AI layer for connected devices.
Co-founder & CRO
Hélène Jaffrelot
CEO
Nicolas Turpault
CTO
Sébastien Vuillemin
Pauline Turpault
Data Engineer
Yoann Pioc
Embedded systems engineer
Fanny Demey
Mobile developer
Eight years of research, productized

Sonaid grew out of eight years of research at INRIA. That work is now a production library, protected by a patent filed at INPI.

We care about your integration

We take the time to understand your device, your constraints, and your market. Your success is how we measure ours.

A dataset you cannot download

Our models learn from a proprietary sound dataset built over eight years. It is not available online, and we keep enriching it.

We fit into your firmware

A model under 2 MB that runs on ESP32, Cortex-M4, and Android. We adapt to your pipeline and your processes.

Licences

We license embedded software so you can ship sound event detection inside your own device.

Integration

Our library runs in your firmware and can send alerts to your platform directly or through API calls.

Device validation

We validate every device, especially the microphone, before licensing, so our software performs at its best on your hardware.

Community

We stay involved in the DCASE research community, giving back and bringing the latest state of the art into our software.

Our Mission

Our mission is simple: give every device the sense of hearing. A device that understands sound can protect, assist, and adapt to the people around it.
‍
It starts with eight years of research at INRIA, now a production library: a patent filed at INPI, a proprietary sound dataset you cannot download, and detection models proven on real devices.

The team is led by a recognized reference in the Sound AI research community, working with PhD-level scientists who still contribute to research. That work runs in production today: vocal distress detection is at TRL 9 on devices in the field.

Sound AI is the wave after Voice AI. Voice gave devices a way to take commands. Sound AI gives them the ability to understand everything else. Sonaid intends to be the standard way a device gains hearing.
Because everyone deserves to feel safe and sound, without compromise.
They awarded us
They back us

We'd love to work with you

Tell us about your use case. We will help you give it the sense of hearing.

Interested to collaborate on sound semantic segmentation ? Get in touch.
Get in touch
HomeAbout usContactpresskit

Copyright © Sonaide

Privacy policyTerms and Conditions