Composable realtime voice

Ox Voice Agent

Bring your prompt, tools, and workflow. Ox handles the realtime voice path from turn detection through speech output.

Ox ownsaudio · turns · ASR · LLM stream · TTS · barge-inYour product keeps telephony, CRM, and sales logic.

01 · Turn detectionSilero VAD + Smart Turn v3.2

02 · TranscriptionQwen3-ASR 0.6B

03 · ReasoningCustomer-selected open model

04 · SpeechQwen3-TTS 1.7B

ReadyConfigure the agent, then turn on your microphone.

Turn on your microphone, wait for the runtime to report Live, then speak normally. You can interrupt it in the middle of a response.

Use headphones for the cleanest full-duplex test.

Browser preview · 16 kHz PCM input · 24 kHz PCM output · use headphones