Multimodal API

This document describes how to use the account-level Multimodal API, which provides various multimodal interface services. Currently, the following service is available:

Text-to-Speech (TTS)

Converts the provided text into a speech audio file.

API Description

POST https://api.ticos.ai/v1/multimodal/tts

Authentication

This endpoint requires Bearer Token authentication. Include your API key or terminal secret in the Authorization header.

Request Headers

Request Body

The request body is in JSON format and contains the following fields:

{
  "text": "Hello, world!",
  "voice": "en-US-Wavenet-D",
  "speed_ratio": 50,
  "pitch_ratio": 50,
  "volume_ratio": 50
}

Request Body Fields

Response

Success Response

• Status Code: 200 OK
• Headers: Content-Type: audio/wav
• Body: The binary WAV audio data.

Error Responses

• 400 Bad Request: Invalid request body (e.g., missing text field).
• 401 Unauthorized: Authentication failed (e.g., invalid token).
• 500 Internal Server Error: Internal server error (e.g., TTS service is not configured or unreachable).

Example Code

cURL

curl -X 'POST' \
  'https://api.ticos.ai/v1/multimodal/tts' \
  -H 'Authorization: Bearer ts_your_api_key_1234567890' \
  -H 'Content-Type: application/json' \
  -d '{
    "text": "This is a test of the TTS service.",
    "voice": "zhifeng_emo"
  }' \
  --output test_audio.wav

Available Voices

Retrieve a list of available voice models for the TTS service.

API Description

GET https://stardust.ticos.cn/tts

Query Parameters