---
title: "Remove Music from Audio Files in Your Browser | ToolSugar"
description: "Separate the voice from the music in an MP3, WAV, M4A, FLAC or OGG file. You get both as MP3s, processed on your device with no upload. Up to 15 minutes."
canonical: https://toolsugar.com/tools/remove-music-from-audio
image: https://toolsugar.com/og.png
---

# Remove Music from Audio

To remove music from an audio file, drop in an MP3, WAV, M4A, FLAC, OGG or Opus recording and an MDX-Net model splits it into a voice file and a music file. Both come back as 192 kbps MP3s. It all runs in your browser with no upload, handles up to 15 minutes per file, and downloads a 66.8 MB model on the first run.

Runs on your device

## How it works

1. Choose an audio file up to 15 minutes long.
2. Press Remove music. The model downloads once and is then kept by your browser.
3. Download the voice MP3, the music MP3, or both.

## Key facts

- **Accepted files**: MP3, WAV, M4A, FLAC, OGG and Opus
- **You get**: A voice MP3 and a music MP3, 192 kbps stereo at 44.1 kHz
- **Length limit**: 15 minutes per file
- **First download**: 66.8 MB model, saved for later runs
- **Speed**: On an Apple M4 Pro: about 16 s per minute of audio with WebGPU, about 3 min 45 s without it

## Common uses

- Take the intro music off a podcast segment that was mixed down too early.
- Rescue a voice memo recorded with a radio playing in the room.
- Hear a lecture recording more clearly when music plays under the speaker.
- Keep the music bed from an old mix when the original session files are gone.

## Questions

### Will the voice sound the same as before?

It stays close, but it is an estimate rather than the original recording. Where music and voice overlap at the same pitch, some detail can go with the music. Reverb on the voice may also fade.

### Can I get lossless output?

This page gives MP3s. The Vocal Remover page offers 16-bit WAV for both stems, and it accepts the same audio files.

### Does the music file contain only instruments?

It holds everything that is not voice, so ambience, effects and background chatter that the model does not treat as voice all land there.

### Why is there a time limit?

Separation is heavy work. In our tests the browser reached nearly 5 GB of memory, and without WebGPU a 15-minute file would take close to an hour even on a fast computer. We have run files up to 15 minutes through WebGPU, so the limit stops at a length we have checked.

[Read the Remove Music from Audio guide](https://toolsugar.com/blog/how-to-remove-music-from-a-video)

## Related tools

- [Remove Music from Video](https://toolsugar.com/tools/remove-music-from-video)
- [Vocal Remover](https://toolsugar.com/tools/vocal-remover)
- [Noise Remover](https://toolsugar.com/tools/noise-remover)
- [Audio Cutter](https://toolsugar.com/tools/audio-cutter)
- [Audio Converter](https://toolsugar.com/tools/audio-converter)