Voice overview

How Leamout models programmable voice, call state, routing, and media control.

Leamout voice separates signaling, routing, durable call state, and application media behind one customer-facing API.

The current voice surface includes:

• voice applications and inbound bindings;
• inbound and outbound durable calls;
• answer and hangup;
• transfer, hold, unhold, playback, stop, recording, and DTMF controls;
• recordings and short-lived playback URLs;
• conferences and participant controls;
• normalized call lifecycle events delivered through the webhook system.

Inbound calls are persisted as durable Leamout resources before application control. The public API addresses those resources while the runtime handles signaling and media execution internally.