Two ways to run Audrio

The same product, delivered either into your own environment or hosted by us. The difference that matters is where the data sits and who holds the keys.

Self-hosted

Everything stays inside your perimeter

Audrio ships as a single container stack you deploy on infrastructure you control. Recordings, transcripts, the search index, the knowledge graph, and every analysis result are stored in your database and your object storage.

Model endpoints are configuration, not architecture. A configuration centre lets an administrator point transcription, analysis, embedding, and re-ranking at self-hosted or EU-region services, so no audio or transcript text leaves the region you choose. That is the mode we recommend for organisations processing recordings of EU residents.

Where data lives: entirely with you. We hold none of it.

Audrio Cloud

We run it, you use it

For smaller teams who would rather not operate infrastructure. Tenants are isolated, and the same pipeline and interface apply. Today the Cloud tier's model endpoints are the non-European defaults. Endpoints are configuration, not architecture, so an instance can be pointed at EU-region services on request, and a one-click EU preset is on the roadmap.

Where data lives: in our hosted environment, which is currently outside the European Union. Until the EU-region build of Cloud is available, recordings of EU residents belong in the self-hosted mode above, not here. We would rather say that plainly than have you discover it during a compliance review.

Commercial multi-tenancy, billing, and tenant self-onboarding are on the roadmap and are not part of this release.

Which one is this demo?

Neither, strictly. This site is a public demonstration running entirely on synthetic sample data, hosted on a general-purpose application platform. It is deliberately separated from every customer environment — the details are here.