Voice Call API Integration Guide
The Voice Chat feature is available to selected customers only.
If you don't have access, please reach out to the Ejento Support Team to request it.
Overview
Step 1 — Call the Generate Token API
Endpoint
POST /response-service/api/v2/agents/{agent_id}/voice-chatPath Parameters
| Parameter | Description |
|---|---|
agent_id | The ID of the agent to dispatch |
Request Body
| Field | Type | Required | Description |
|---|---|---|---|
query_source | string | ✅ Yes | Source of the voice call request (e.g., "web") |
history | array | ❌ No | Optional list of prior conversation turns |
chat_thread_id | number | ❌ No | Optional chat thread ID to continue an existing session |
Example Request
POST /response-service/api/v2/agents/{agent_id}/voice-chat
Content-Type: application/json
{
"chat_thread_id": 1,
"history": [
{
"bot": "Hi there!",
"user": "Hello"
}
],
"query_source": "web"
}Example Response
{
"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
}The API dispatches your AI agent to a LiveKit room and returns everything your frontend needs to join the same room as a participant — the server URL and a pre-signed participant token.
Step 2 — Connect Your Frontend to LiveKit
livekit_url and participant_token from the API response, use them to connect your frontend to the LiveKit room.Room object to connect using your URL and token, handle participant events, publish/subscribe to audio tracks, and manage disconnection scenarios.Step 3 — Build & Test Your Frontend
Running the Playground Locally
livekit_url and participant_token, then:livekit_url into the wss://url fieldparticipant_token into the room token fieldOnce connected, the playground opens a live session view showing the Agent Audio visualizer and a Chat panel. Your agent will join the room automatically and you can start speaking with it in real time.