SJinn
Seed Audio 1.0

Seed Audio 1.0

Create Seed Audio 1.0 audio generation tasks through the SJinn REST API

Tool Overview

The Seed Audio 1.0 tool creates audio from a text prompt. It can also use either reference audio for voice or style guidance, or one reference image for mood and scene guidance.

Tool Identifier

seed-audio-1-0-api

Parameters

Required Parameters

prompt

  • Type: string
  • Description: Text description of the audio to generate
  • Validation: Must be a non-empty string
  • Example: "A calm female voice reading a bedtime story"

Optional Parameters

audio_urls

  • Type: string[]
  • Default: []
  • Limit: Up to 3 audio URLs
  • Description: Reference audio used to guide voice, tone, style, or energy
  • Prompt References: Cite reference audio as @Audio1, @Audio2, and @Audio3 in the same order as audio_urls

image_urls

  • Type: string[]
  • Default: []
  • Limit: Up to 1 image URL
  • Description: Reference image used to guide mood, scene, or atmosphere

Constraints

  • Use either audio_urls or image_urls, not both.
  • All media URLs must be complete HTTP/HTTPS addresses.
  • Relative paths are not supported.

Pricing

  • Credits Consumed: 100 credits per task

Request Examples

Text-to-Audio

curl -X POST https://sjinn.ai/api/un-api/create_tool_task \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "tool_type": "seed-audio-1-0-api",
    "input": {
      "prompt": "A calm female voice reading a bedtime story"
    }
  }'

Reference Audio

curl -X POST https://sjinn.ai/api/un-api/create_tool_task \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "tool_type": "seed-audio-1-0-api",
    "input": {
      "prompt": "Narrate in the style of @Audio1 with the energy of @Audio2",
      "audio_urls": [
        "https://example.com/reference-voice.mp3",
        "https://example.com/reference-energy.wav"
      ]
    }
  }'

Reference Image

curl -X POST https://sjinn.ai/api/un-api/create_tool_task \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "tool_type": "seed-audio-1-0-api",
    "input": {
      "prompt": "Background music matching the mood of this scene",
      "image_urls": ["https://example.com/scene.png"]
    }
  }'

Using JavaScript

const response = await fetch('https://sjinn.ai/api/un-api/create_tool_task', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    tool_type: 'seed-audio-1-0-api',
    input: {
      prompt: 'A calm female voice reading a bedtime story',
      audio_urls: ['https://example.com/reference-voice.mp3'],
    },
  }),
});

const result = await response.json();
console.log('Task ID:', result.data.task_id);

Using Python

import requests

url = 'https://sjinn.ai/api/un-api/create_tool_task'
headers = {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json',
}
data = {
    'tool_type': 'seed-audio-1-0-api',
    'input': {
        'prompt': 'A calm female voice reading a bedtime story',
        'image_urls': ['https://example.com/scene.png'],
    },
}

response = requests.post(url, json=data, headers=headers)
result = response.json()
print('Task ID:', result['data']['task_id'])

Response Examples

Success Response

{
  "success": true,
  "errorMsg": "",
  "error_code": 0,
  "data": {
    "task_id": "550e8400-e29b-41d4-a716-446655440000"
  }
}

Error Response

{
  "success": false,
  "errorMsg": "audio_urls maximum is 3",
  "error_code": 400
}

{
  "success": false,
  "errorMsg": "Reference audio and reference image cannot be used together. Please use only one.",
  "error_code": 101
}

Best Practices

  1. Keep prompts specific about speaker, tone, pacing, music style, or sound design.
  2. Use audio_urls when voice or audio style matters, and cite them with @Audio1, @Audio2, or @Audio3.
  3. Use image_urls when the desired audio should match a visual mood or scene.
  4. Poll task status with query_tool_task_status until the task completes or fails.

Related Links