Remove Vocals from a Song

Why this matters

Classic vocal removal inverted phase and zeroed the center — it also gutted the bass and drums. Model-based separation keeps the music intact while removing the voice.

Whether it is karaoke night, a remix, or a practice track, you want the instrumental with full frequency range and stereo image.

How it works

Submit the song with stems=[instrumental]. The API returns the instrumental stem (and the vocal stem if you ask) with no phase cancellation.

Submit a job with an input URL (S3, GCS or HTTPS), poll the job URL, and download the rendered output. No GPU, no queues, no ffmpeg builds to babysit.

Code

import requests, time

API = "https://api.mlslabs.io/v1/vocal-separator/jobs"
headers = {"X-API-Key": "YOUR_API_KEY"}
payload = {"input": "s3://bucket/song.mp3", "stems": ["instrumental"]}

resp = requests.post(API, json=payload, headers=headers)
job = resp.json()
while job["status"] not in ("succeeded", "failed"):
    time.sleep(3)
    job = requests.get(job["url"], headers=headers).json()

print("Output:", job["output_url"])

Pro tips

• Use lossless input (WAV/FLAC) when available — MP3 lossy artifacts separate slightly worse.
• For remixes, request both stems and re-mix at will.
• Check the low end on the instrumental: some presets preserve bass better than others.

Try vocal removal in Postman — sample song prefilled.