Two boxes, no call
Box 1: WebRTC, media in the browser, from the W3C Recommendation.
Box 2: SIP, signaling, from RFC 3261.
A note: xAI's page also names WebSocket and LiveKit as ways to connect.
Stop. Do not open a socket and do not buy a number.
Sources and go deeper
- W3C, WebRTC Recommendation, 13 March 2025 spec last checked 2026-10
- WebRTC — Wikipedia wikipedia last checked 2026-10
- RFC 3261, SIP: Session Initiation Protocol spec last checked 2026-10
- Speech to Speech, docs.x.ai vendor The page title is Speech to Speech. The body names the Grok Speech to Speech API and also calls the site SpaceXSI Docs. last checked 2026-10