Update workspace voice integration

PATCH

https://app.lumenia.net/api/v1/workspaces/:workspaceId/integrations/voice

PATCH

/api/v1/workspaces/:workspaceId/integrations/voice

$ curl -X PATCH https://app.lumenia.net/api/v1/workspaces/workspaceId/integrations/voice \
>      -H "X-Account-Id: X-Account-Id" \
>      -H "Authorization: Bearer <API_KEY>" \
>      -H "Content-Type: application/json" \
>      -d '{
>   "prompt": "You are a concise scheduling assistant for a dental clinic."
> }'

1 {
2   "voice": {
3     "connected": true,
4     "fallbackTranscribers": [],
5     "fallbackVoices": [],
6     "firstMessage": "Hello! How can I help you today?",
7     "phoneNumber": "+14155551234",
8     "prompt": "You are a concise scheduling assistant for a dental clinic.",
9     "promptSource": "manual",
10     "promptTemplate": null,
11     "transcriber": {
12       "language": "en-US",
13       "model": "nova-2-phonecall",
14       "provider": "deepgram"
15     },
16     "voice": {
17       "model": null,
18       "provider": "vapi",
19       "voiceId": "Kai"
20     }
21   }
22 }

Update selected voice integration settings.

Authentication

AuthorizationBearer

Bearer authentication of the form Bearer <API_KEY>, where API_KEY is your API key.

Path parameters

workspaceIdstringRequired

Workspace whose voice integration to update.

Request

Settings to update. Omitted fields remain unchanged. Use the options endpoint for supported values. The routing number cannot be changed here.

firstMessagestringOptional

The greeting the assistant speaks first.

promptstringOptional

System prompt for the live voice conversation.

voiceobjectOptional

The text-to-speech voice. For the vapi provider set model to null or omit it; for 11labs both voiceId and model are required.

fallbackVoiceslist of objectsOptional

Replaces the fallback voice chain. Each entry needs a provider; send an empty array to clear.

transcriberobjectOptional

The speech-to-text transcriber.

fallbackTranscriberslist of objectsOptional

Replaces the fallback transcriber chain. Send an empty array to clear.

Response

The updated voice integration.

voiceobject

The assistant config after the update — see GET …/integrations/voice for the field notes.

Errors

400

Bad Request Error

403

Forbidden Error

404

Not Found Error

500

Internal Server Error

Authentication

Path parameters

Headers

Request

Response

Errors