vui speaks and listens entirely on-device — no server, no account needed to talk, nothing leaving your iPhone. Free while it's in beta.
iPhone with iOS 26+ · one-time model download · free during beta
Not on an iPhone? Talk to it in the browser — no account needed.
35 seconds, sound on — every word is generated on the phone.
Real voices
Natural, expressive speech — hesitations, breaths and all — from a 300M model running on the iPhone GPU.
Talk to it
Hold the mic and get a spoken reply. Recognition, the language model and the voice all run on the device.
Private by construction
Your words never leave the phone. There is no server doing the thinking, because there is no server.
Works offline
On a plane, on the tube, off-grid. Once the models are downloaded, vui needs no connection.
An AI receptionist for your business phone. Hook up your number and it answers every call — books jobs, takes orders and answers questions, day or night.
See the receptionistWe own the whole stack — recognition, reasoning and voice — so the API is one bill with nothing stacked on top and no keys to bring.
Voice agents
Live conversation, all-in — recognition, reasoning and voice. No component stacking, no keys to bring.
Transcription
akro-v1 speech-to-text with diarization, billed by the second ($0.01 minimum per request). Cheaper than OpenAI, Deepgram and AssemblyAI.
Speech
Expressive text-to-speech with voice cloning — about $0.45 per hour of audio.
The beta is open. Bring an Apple ID and a few minutes for the model download.
Get the betaor come say hello on Discord