Listen
Voice classifiers your users help train
Ship custom voice commands to phones, browsers, Raspberry Pis. Real-time adaptation per user — no cloud round-trip.
See the SDK →voice-controlled AI that runs offline
Learns your users. Ships without cloud bills.
Three products. One adaptive engine.
Sentinel
Offline acoustic sensors for industrial fleets
Always-on factory-floor monitoring. Encrypted bundles, signed expiry, no phone-home. Hardware reference designs included.
Join the waitlist →Vision
Image recognition that learns from your data
Same adaptive engine, applied to images. Per-user classifiers that learn from a handful of examples.
Add voice to your product in 3 lines.
from rivlet_listen import Model
model = Model.load("doorbell.rivlet")
print(model.predict(audio_wav_bytes))Python, Rust, C, JavaScript, or a standalone binary. Ships everywhere from Raspberry Pi to the browser.
One free tier. Two paid. Custom for the rest.
Start free, upgrade when you ship.
Hobby
Try every model in your browser. Always free.
Free
- 1 deployed model
- ≤ 5 user-added words per model
- WASM bundle only (browser embed)
Developer
Ship one product to real users.
$49/mo
Recommended
- 10 deployed models
- ≤ 50 words per model
- Python wheel + WASM bundle
Startup
Embed in a real product, with real users.
$499/mo
- Unlimited models
- ≤ 200 words per model
- All bindings (Python, C, Rust, WASM, binary)
Need more? Enterprise plans available — talk to us.