Generate Token for Voice Call

POST

/response-service/api/v2/agents/{agent_id}/voice-chat

Dispatches an agent to a LiveKit room and generates a participant token for the user to join the same room.

New to this API? Check out the Voice Call API Integration Guide for a complete walkthrough — from calling this endpoint to building and testing your frontend with LiveKit.

Looking for a higher-level overview? See Voice Calling with Assistants to understand how voice calling fits into the Ejento AI ecosystem.

Path Parameters

agent_id: ID of the agent to dispatch.

Request Body

history: Optional list of prior conversation turns

chat_thread_id: Optional chat thread ID

query_source: Required source of the voice call request (e.g., 'web')

Returns

JSON Response with participant token, LiveKit URL, and the chat_thread_id used for this call.

Current Version

2.0.0

Request

Path Params

Header Params

Body Params application/jsonRequired

Examples

Responses

🟢200

application/json

Successfully dispatched agent and generated token

Body

🟠400

🟠422

🔴500

Request Example

Shell

JavaScript

Java

Swift

curl --location -g --request POST 'https:///response-service/api/v2/agents//voice-chat' \
--header 'Ocp-Apim-Subscription-Key: ' \
--header 'Authorization: Bearer ' \
--header 'Content-Type: application/json' \
--data-raw '{
    "chat_thread_id": 1,
    "history": [
        {
            "bot": "Hi there!",
            "user": "Hello"
        }
    ],
    "query_source": "web"
}'

Response Example

200 - Example 1

{
    "data": {
        "chat_thread_id": 123,
        "livekit_url": "wss://your-livekit-server.com",
        "participant_token": "your-participant-token"
    },
    "message": "Successfully dispatched agent and generated participant token",
    "success": true
}

Generate Agent Response (Non-Streaming)

Get an Agent Response