How licensing works
- 01
Tell us your hardware. Target chipset, task (transcription, affect, or both), and expected fleet size. A person replies within one business day.
- 02
Free evaluation build. A compiled build for your target device under a time-boxed license — no card, no commitment. Accuracy and latency are measured on your own audio, on your own silicon.
- 03
License per device. If the numbers hold, production is a flat annual per-device rate with model updates included. Custom builds for your footprint and task are scoped separately.
Free evaluation. No card, no commitment.
Why a lab, not a model zoo
The model family
In the historical August 2026 speech-emotion-bench release, oruk Resonance 1 reached 77.8% seven-class accuracy. Oruk trained in-distribution while the comparison systems were evaluated zero-shot. Full and subset evaluations used different audio denominators; the methodology page documents them. This is not an accuracy claim for the current API, Orukeet, or a device build. Those require their own measurements.
The alternative
Many free runtimes are designed around transcript-only prototypes. An evaluated commercial program can add acoustic emotion and speaking-style outputs, tuning for your microphone channel and acoustics, acceptance criteria on your hardware, and explicit support and survival terms.
Free evaluation. No card, no commitment.
Exactly what you are buying
The limits below are stated here so you do not discover them after integrating.
- On-device language scope
- The file models described here process English speech. Hosted multilingual streaming is available separately through the Realtime preview; on-device availability and language support are agreed with your team.
- Specs are measured, not quoted
- We do not publish generic size, memory, power, or latency figures. Those depend on your chipset and task, and are established during evaluation on your hardware.
- Historical speech-emotion benchmark
- The 77.8% speech-emotion-bench result belongs to oruk Resonance 1 in the historical August 2026 release (methodology, including the in-distribution training caveat). It does not measure the current hosted API, Orukeet, or a proposed device build. We evaluate a device build against your acceptance criteria on your hardware.
- What emotion scores are — and are not
- Model annotations of how speech sounds — not intent, truth, medical, employment, or psychological judgments. See responsible use.
- Data and training
- On-device inference means we never see your audio at all. Across the whole product line, oruk never trains on customer content.
- Prefer hosted?
- The hosted model family runs behind a synchronous REST API on plans from $9/month with a 7-day free trial — the fastest way to validate accuracy before committing to a hardware evaluation. See the cloud API.
Free evaluation. No card, no commitment.
Prove it on your silicon
Send your target hardware, task, and expected fleet size. You get a compiled evaluation build and measured numbers on your own audio — then decide.