Create a call with inline configuration

Run a call without a saved agent — the whole configuration travels in the request. Nothing is stored and nothing is reused.

Use it for per-user personas, generated agents, or testing a prompt without leaving a record behind. If the same configuration is used repeatedly, create a real agent instead: you gain version history, tool assignment, a phone number, corpora, and the dashboard.

POST /calls/inline

Create a call with inline configuration

cURL

curl --request POST \
  --url https://api.omnia-voice.com/api/v1/calls/inline \
  --header 'Content-Type: application/json' \
  --header 'X-API-Key: <api-key>' \
  --data '
{
  "systemPrompt": "<string>",
  "voice": "Mark",
  "language": "en"
}
'

201

{
  "id": "<string>",
  "websocketUrl": "<string>",
  "status": "<string>",
  "connectionType": "<string>",
  "createdAt": "2023-11-07T05:31:56Z"
}

Authorizations

X-API-Key · string · header · required

API key for authentication

Body

application/json

systemPrompt · string · required

The entire prompt. There is no basePrompt/context/ customInstructions layering here — that belongs to saved agents.

voice · string · required

Voice name or ID. See GET /voices.

Example:

"Mark"

language · string · required

Language code. See GET /languages.

Example:

"en"

model · enum<string> · default: llama

Available options: gemma, llama, glm

temperature · number · default: 0

Required range: 0 <= x <= 1

greeting · string

Appended to the prompt as the opening line when the agent speaks first. Ignored when firstSpeaker is user.

firstSpeaker · enum<string> · default: agent

Available options: agent, user

interruptible · boolean · default: false

Applies to the agent's first message only.

maxDuration · integer · default: 3000

Hard cap in seconds.

Required range: 60 <= x <= 7200

inactivityTimeout · integer

Seconds of silence before the agent reacts.

Required range: 10 <= x <= 300

inactivityAction · enum<string>

Available options: hang_up, prompt

recordingEnabled · boolean · default: true

connectionType · enum<string> · default: twilio

Which transport the returned websocketUrl is prepared for.

twilio / telnyx / plivo — carrier media streams, G.711 audio. websocket — your own client; you choose the sample rates and codec. webrtc — browser transport, normally reached through the SDK.

Any other value returns 400.

Available options: twilio, telnyx, plivo, websocket, webrtc

websocket · object

Audio parameters. Only read when connectionType is websocket — telephony transports force G.711 at 8000 Hz regardless of what is sent here.

Show child attributes

selectedTools · object[]

Show child attributes

metadata · object

Response

201

application/json

Call session created

id · string

websocketUrl · string

Short-lived, single-use URL to stream audio to.

status · string

connectionType · string

createdAt · string<date-time>