AI - オーディオ・ステム・ジェネレーター
Split music files into layers with AI


Select all audio stems you want to generate:

Noise Cancelling

✔ Usage based billing
✔ Commercial rights
✔ No subscription required

What are audio stems?

Music stems are audio files that break down a complete track into individual mixes or instruments. This tool can extract voices and separate them, as well as instruments such as piano, drums, bass, guitar, synthesizer, and string and wind instruments into single stems.

Stems are widely used by artists for more flexibility in editing and remixes. Stem files enable the addition of track parts that best align with your video, allowing for a gradual build-up of music starting with just the drums. Stems also allow for customization of tracks by removing unwanted instruments while keeping the rest intact.

Instrumental Stem

The instrumental audio stem file will contain everything except the vocals. This stem is ideal for use as a karaoke or instrumental version of a song.

Vocals Stem

This stem contains the extracted vocals and voices in a song. Use this stem if you want to extract vocals from a piece of music.

Voice Stem

The voice stem contains clean voices extracted from an audio file. Use it when you want to clean up an audio recording of a noisy street, crowd, or windy forest with irritating background noise.

Piano Stem

This stem contains pianos.

Drums Stem

This stem contains drums.

Guitar Stem

The tool distinguishes between two types of instrument stems. You can choose between electric and acoustic guitars.

Synthesizer Stem

This stem contains electronic musical instruments and synthesizer.

Strings Stem

This stem contains strings like violin, cello and fiddle.

Wind Instrument Stem

This audio file contains wind instruments in the categories of brass instruments (such as horns, trumpets, trombones, euphoniums, and tubas) and woodwind instruments (such as recorders, flutes, oboes, clarinets, saxophones, and bassoons).

Noise Stem

This stem contains everything that is not recognized as a voice.


Batch Processing

You can upload multiple files to batch process a bunch of clips.

De-echo voice

This option allows you to reduce the amount of echo from a voice.

Algorithm & Filter

To filter the voice, select your algorithm and its intensity.

Example 1: Voice Stem

A short conversation between a guy and a girl speaking in Indonesian language with music in the background.
Algorithm: Orion v4, Filter: normal




Stem 1: Voice

Stem 2: Noise


Example 2: Voice Stem

A distant conversation in a large room.
Algorithm: Orion v4, Filter: normal




Stem 1: Voice

Stem 2: Noise


What should I bear in mind when generating stems?

It is important to obtain permission before creating or using stems for which you do not own the copyright. Permission from the label or composer may be required to use stems in the production of commercial products because they change the overall impression of a song.

How long does it take to generate audio stems?

The generations does take only some minutes. More stems need more time. You will get an email as soon your files are ready to download.

Supported File formats

.mp3, .wav, .ogg, .flac, .aiff, .acc, .mp4, .mkv, .m4a
Size Limit: 2GB


This AI tool is pay-per-use, so you don't need a subscription to use it and you only pay for what you need. This way, our tool can be cheaper if you are only working on a small project or just want to run a few tests without subscribing to a service.

For every 20 seconds of audio to be processed, we charge 1 AI credit, which is about $0.08. For example, creating a voice stem from a three-minute song costs just $0.72.

To better predict the total price of your project, you can upload your file and select the stems you need to calculate the cost.

Image credits: Rodrigo Souza, Mido Makasardi, RDNE Stock project, ALTEREDSNAPS, Los Muertos Crew


AI Tools Tools MP3 Music WAV Premium Tools Adobe Premiere Pro Apple Final Cut Pro X Avid Media Composer DaVinci Resolve Tools for Audio Workflows