Voice Home is a research preview of where the Annotiq Data Engine could go next — an ambient assistant trained on the same labeled ground truth that powers our customers’ models.
Our main offering is data labeling for AI teams. Voice Home is not generally available, has no shipping date, and is shown here to make the company’s direction concrete. If you’d like to be notified when there’s news, join the list below.
On-device wake-word detection with sub-100ms latency. Never sends raw audio off the device unless you ask it to.
Voice fingerprints stored locally. End-to-end encrypted sync. You own every transcript; nothing trains a third-party model.
Routines emerge from your patterns, not a marketing taxonomy. Replace placeholder copy with real product description.
The labeling stack behind everything we ship — generative AI, physical AI, autonomy, RL environments, and yes, eventually Voice Home too.