Chat Mode

Chat Mode provides standard text and voice chat functionality with pre-recorded avatar responses. Messages are sent via WebSocket and responses are delivered as text with optional audio/video URLs.

Overview

Workflow:

1. Establish WebSocket connection (see WebSocket API)
2. Send text or audio messages with isLive: false
3. Receive text response with optional fileUrl for audio/video playback

tip:  

For real-time interactive sessions with streaming 3D avatar responses, use Live Mode instead.

Sending Text Messages

Send a text message via WebSocket.

WebSocket Message:

{
  "avatar_id": "avatar-id-123",
  "user_id": "user-123",
  "chat_type": "text",
  "language": "en",
  "request": "Hello, how are you?",
  "requestType": "text",
  "source": "my-source-app-name",
  "isLive": false,
  "session": "session-identifier"
}

Field	Type	Required	Description
avatar_id	string	Yes	Target avatar identifier
user_id	string	Yes	User identifier
chat_type	string	Yes	"text"
language	string	Yes	Language code (e.g., "en")
request	string	Yes	Message text content
requestType	string	Yes	"text"
source	string	Yes	Client identifier (e.g., "mobile-app")
isLive	boolean	Yes	false for Chat Mode
session	string	No	Session identifier from configuration

Sending Audio Messages

Send an audio message via WebSocket.

WebSocket Message:

{
  "avatar_id": "avatar-id-123",
  "user_id": "user-123",
  "chat_type": "voice",
  "language": "en",
  "requestType": "audio",
  "source": "my-source-app-name",
  "isLive": false,
  "isPushToTalk": true,
  "file_base64_data": "UklGRiQAAABXQVZFZm10IBAAAAABAAEA...",
  "session": "session-identifier"
}

Field	Type	Required	Description
avatar_id	string	Yes	Target avatar identifier
user_id	string	Yes	User identifier
chat_type	string	Yes	"voice"
language	string	Yes	Language code
requestType	string	Yes	"audio"
source	string	Yes	Client identifier
isLive	boolean	Yes	false for Chat Mode
isPushToTalk	boolean	No	true when VAD is disabled (see Voice Activity Detection)
file_base64_data	string	Yes	Base64-encoded audio blob
session	string	No	Session identifier

Standard Audio Format

For Chat Mode audio messages, use the following audio format:

Property	Value
Format	WAV or WebM
Sample Rate	16,000 Hz (16 kHz)
Channels	Mono (1 channel)
Encoding	Base64
Transmission Field	file_base64_data

Audio Recording Constraints:

{
  "sampleRate": 16000,
  "echoCancellation": true,
  "noiseSuppression": true,
  "autoGainControl": true,
  "channelCount": 1
}

note:  

For Live Mode audio using PCM format, see the Live Mode documentation. For a comparison of both audio formats, see Audio Format Comparison.