One telephony control plane
Bring inbound PSTN, outbound dialing, browser rooms, and call recording into the same MediaSFU workflow instead of stitching together separate stacks.
MediaSFU telephony is not just SIP registration plus a dial tone. It is the layer that lets one room coordinate callers, AI voice agents, human operators, IVR prompts, recordings, and backend actions in a single flow. If your product needs phone numbers, escalation logic, or AI-assisted voice experiences, this guide is where the real operating model starts.
Bring inbound PSTN, outbound dialing, browser rooms, and call recording into the same MediaSFU workflow instead of stitching together separate stacks.
Answer with AI voice agents, play IVR prompts, then hand off to a human or operator queue without moving the caller to another system.
Control prompts, callbacks, routing, webhooks, and provider-specific SIP behavior from the same room-aware configuration surface.
Production teams use this surface for inbound support lines, AI receptionists, callback queues, operator assist, outbound campaigns, and hybrid voice workflows where PSTN callers and browser users need to share the same MediaSFU logic.
Provider credentials matter, but production telephony is the larger sequence: carrier entry, room attach, AI or IVR routing, human escalation, and audit capture. The best SIP setups are designed around that full path.
A SIP INVITE or DID is only the first mile. Once attached, the call participates in the same MediaSFU room and routing model as your AI and operator workflows.
Prompt playback, transcription, model responses, and backend actions should be designed as one path instead of separate disconnected tools.
If the caller needs a person, pass intent, transcript summary, and prior actions so the operator continues the job instead of restarting discovery.
Track routing decisions, AI behavior, callback state, and escalation outcomes so support and operations teams can explain every call path.
Watch MediaSFU keep PSTN callers, WebRTC participants, AI routing, and recording policy in the same operational room.
See IVR prompts, AI handling, and human escalation stay in one stack rather than bouncing the caller across disconnected systems.
Start with the ready-made MediaSFU telephony apps if you want working mobile, desktop, or web call experiences before deep SIP customization.
Use Start Here if you need the overall flow for providers, credentials, DIDs, and whether a basic or detailed setup path fits your team.
Go straight to Critical Setup when you are managing trunks, extra fields, routing behavior, AI handoff rules, or provider-specific configuration details.
If you need working telephony apps before you tune provider fields, MediaSFU already ships production-oriented VOIP surfaces for mobile, desktop, and web teams.
The rest of this guide walks from provider and trunk setup into DIDs, critical extra fields, custom audio assets, AI-driven call behavior, and the ready-made applications that let you move faster when you do not want to build the full telephony shell yourself.
