This endpoint allows users to separate audio stems from a provided track using AI.


Stem Separation API Overview

The Stem Separation API by Soundverse allows you to extract specific parts (or “stems”) of a song such as:

  • Vocals

  • Instruments

  • All individual stems (vocals, drums, bass, etc.)

This is useful for remixing, karaoke creation, music analysis, or background score extraction.


Core Parameters

🛈 The parameters below define the request body for this API.

ParameterTypeRequired?Description
audioUrlstring✅ RequiredURL to the input audio file.

Request Body

Request body: 
{
  "audioUrl": "string"
}

Code Samples

  • Python code sample
  import requests

  url = "https://api.soundverse.ai/v1/generate/stem-separation/vocals"
  headers = {
      "Authorization": "Bearer your_api_key_here",
      "Content-Type": "application/json"
  }
  payload = {
    "audioUrl": "string"
  }

  response = requests.post(url, json=payload, headers=headers)
  print(response.json())
  • cURL
    curl -X POST "https://api.soundverse.ai/v1/generate/stem-separation/vocals" \
    -H "Authorization: Bearer your_api_key_here" \
    -H "Content-Type: application/json" \
    -d '{
    "audioUrl": "string"
    }'

Response

[
  {
  "audioName": "vocals",
  "audioUrl": "str"
  }
]

Possible Errors

  • If NSFW words are detected

{
"success": False,
"message": "NSFW words detected. Please try again with a different prompt.",
}

  • If Artist name is detected

{
"success": False,
"message": "Artist name detected. Please try again with a different prompt.",
}

  • Generation Fail Error

{
"Detail": "str"
}