Skip to content

Try Resonance 2 early.

Tell us a little about yourself and what you’re building.

A brief description of your use case.

We’ll use these details to review your application and contact you about access. Privacy policy

Get started free

Explore oruk

A research lab for speech AI

We’re researchers training speech AI to understand words and how they’re spoken. Our models use tone, rhythm, and emphasis to return transcripts with emotion and speaking-style labels. We publish our evaluations so you can inspect the results before building with them.

Backed by a16z speedrun.

Explore our modelsRead the research

The lab also supports the oruk organization, our nonprofit arm bringing speech technology to endangered languages.

The team

Advisors

What we believe

  • Words and delivery

    We model transcription, emotion, and speaking style together, so applications can work with what was said and how it sounded.

  • Published evaluations

    We publish evaluation methods, comparison results, and limitations so researchers and developers can inspect our work.

  • From research to your product

    Use our models through one API, with SDKs, examples, and a playground for your own audio.