---
title: "Audio Stem Splitter — Documentation & FAQ"
description: "How to use Audio Stem Splitter, plus answers to common questions."
canonical: "https://www.razi.pro/tools/vocal-separator/docs"
tags: ["audio stem splitter", "acapella extractor", "vocal separator", "stem separation", "vocal isolation", "extract vocals for remix", "separate audio stems", "demucs ai audio"]
source: "razi.pro"
---

# Audio Stem Splitter — Documentation & FAQ

Documentation for [Audio Stem Splitter](https://www.razi.pro/tools/vocal-separator) — Isolate vocals, extract clean acapella stems, and separate backing tracks for music production, remixes, and video editing using Meta's Demucs AI model.

## Frequently asked questions

### How do I remove vocals from a song?

Upload your audio file (MP3, WAV, FLAC) or video file (MP4, MOV), select your separation mode (Vocals, Instrumental, or Both), then click 'Separate Vocals'. The AI will process the audio and provide separate downloads.

### What AI model is used for vocal separation?

The tool uses Meta's Demucs model, an open-source deep learning model specifically trained for music source separation. It separates audio into vocals and instrumentals with high fidelity.

### How long does vocal separation take?

Processing typically takes 2-5 minutes depending on the length of the song. Longer tracks take more time. The AI runs server-side on a dedicated processing server.

### What is the output quality?

Output is in WAV format for maximum quality. WAV is lossless, so the separated vocals and instrumentals maintain the highest possible audio fidelity.

### Can I use this for karaoke?

Yes! Select 'Instrumental Only' mode to get a karaoke version of any song with vocals removed. The AI does an excellent job separating vocals from most studio recordings.

### What is the maximum file size?

Audio is uploaded through the request body, which is capped at a few megabytes. Trim longer recordings before uploading.