Split audio into stems
May 15, 2025 (November 16, 2026)
Table of contents
Upload an audio file (MP3, WAV, M4A, FLAC) and split it into separate instrument stems (vocals, drums, bass, etc).
POST raw content using Make.com and similar nocode tools.
https://api.useapi.net/v1/tempolor/music/stems-splitter/?…
Request Headers
Authorization: Bearer {API token}
Content-Type: select from the table below
API tokenis required, see Setup useapi.net for details.Content-Typeis required, select from the table below:
| Content-Type | File Extension |
|---|---|
| audio/mpeg | mp3 |
| audio/wav | wav |
| audio/wave | wav |
| audio/mp4 | m4a |
| audio/x-m4a | m4a |
| audio/flac | flac |
| audio/x-flac | flac |
Query Parameters
user_idis optional when only one account configured. However, if you have multiple accounts configured, this parameter becomes required.
Responses
-
When the response includes
stemsZipUrl, the processing is complete and you can download the ZIP file containing the separated stems.{ "state": 2, "title": "audio_file.mp3", "stemsZipUrl": "https://example.com/path/to/stems.zip" } -
The server encountered an error processing your request.
{ "error": "<Error message>", "code": 400 } -
{ "error": "Unauthorized", "code": 401 } -
596 Account signed out
TemPolor signed the account out, so its session ticket no longer works. The API stops using the account until you add it again with fresh credentials at POST accounts, see Setup TemPolor. Requests without
user_iduse your other accounts in the meantime.{ "error": "TemPolor signed out account 101441809 (You are currently logged out (400006)). Add it again with POST /v1/tempolor/accounts using fresh credentials, see https://useapi.net/docs/start-here/setup-tempolor", "code": 596 }If every account you have is in this state, a request without
user_idreturns 596 as well.
Model
The split result, with stemsZipUrl linking to a ZIP of the separated stems once processing is complete.
The split result, field by field
{ // TypeScript, all fields are optional
state: number
title: string
stemsZipUrl: string // URL to download the zip with separated stems
audioType: string
detail: string[]
error: string
code: number
}
Examples
-
curl "https://api.useapi.net/v1/tempolor/music/stems-splitter/?user_id=<user_id>" \ -H "Authorization: Bearer …" \ -H "Content-Type: audio/mpeg" \ --data-binary @/path/to/your/audio-track.mp3 -
const token = "API token"; const user_id = "Previously configured account"; const apiUrl = `https://api.useapi.net/v1/tempolor/music/stems-splitter/?user_id=${user_id}`; let blob; /* // Example 1: Fetch audio from URL const audioUrl = "https://example.com/path/to/audio-track.mp3"; const responseAudio = await fetch(audioUrl); blob = await responseAudio.blob(); */ /* // Example 2: Load audio from local file (Blob) const fsp = require('fs').promises; const audioFileName = "./audio-track.mp3"; blob = new Blob([await fsp.readFile(audioFileName)]); */ /* // Example 3: Load from input file html element // <input id="audio-file" type="file"> const audioFile = document.getElementById(`audio-file`); if (audioFile.files[0]) blob = audioFile.files[0]; */ // Upload the audio file for stem splitting const response = await fetch(apiUrl, { method: "POST", headers: { "Authorization": `Bearer ${token}`, "Content-Type": "audio/mpeg" }, body: blob }); const result = await response.json(); console.log("response", {response, result}); -
import requests import time token = "API token" user_id = "Previously configured account" apiUrl = f"https://api.useapi.net/v1/tempolor/music/stems-splitter/?user_id={user_id}" headers = { 'Authorization': f'Bearer {token}', 'Content-Type': 'audio/mpeg' } # # Example 1: Fetch audio from URL # import requests # audio_url = "https://example.com/path/to/audio-track.mp3" # response_audio = requests.get(audio_url) # file_content = response_audio.content # # Example 2: Load audio from local file audio_file_path = "./audio-track.mp3" with open(audio_file_path, 'rb') as audio_file: file_content = audio_file.read() response = requests.post(apiUrl, headers=headers, data=file_content) result = response.json() print(response.status_code, result)