AudioStem.ai cloud · beta
Write it with AI or split it. Practice, mix, master.
Sign in
VOCAL REMOVER

Remove vocals from a song for karaoke or an acapella

Upload a song and AI pulls the singing apart from the band. Mute the vocal stem and you have an instrumental to sing over; solo it and you have the acapella on its own. Both come out of the same separation and play in sync in your browser.

The result opens in a full mixer instead of coming back as two loose files, so you can also loop the verse you are rehearsing, turn the original singer down rather than off as a guide, and on Studio move the song into your key before you export it.

Start free, no card required
AudioStem.ai with separated stems behind the HD full mix dialog, offering Faithful, Studio Master, Wide/3D and Warm Tape versions of a backing track.AudioStem.ai with separated stems behind the HD full mix dialog, offering Faithful, Studio Master, Wide/3D and Warm Tape versions of a backing track.

What you can do

INSTRUMENTAL

A karaoke track from any song

Mute the vocals stem and the rest of the recording keeps playing: drums, bass and every other instrument, still locked together. That is a backing track you can sing over straight away.

ACAPELLA

Keep only the voice

Solo the vocals stem to hear the singer with nothing around it. Remixers start from the isolated vocal; singers use it to study phrasing, breaths and ad libs.

GUIDE VOCAL

Turn the singer down, not off

Every stem has its own volume fader, so the original voice can stay quietly underneath while you learn the melody. Mute it once the part is yours.

VOCAL MODELS

Dedicated voice and band splits

Studio’s full catalog adds the Karaoke (vocals / instrumental) option and the Premium vocals model. Both return just two stems, a vocal and an instrumental, for songs where the voice is the only thing you need apart.

YOUR KEY

Move the backing track into your range

On Studio, key control transposes the whole song across 12 semitones, and the custom mix you download keeps the key you picked.

EXPORT

Take the version you built

Paid plans download the stems as MP3. Studio adds lossless WAV and a custom mix made from your levels, mutes, edits, EQ and key.

How it works

  1. 1

    Add the song

    Upload an audio file or the sound from a video, or paste a YouTube link; each plan comes with a YouTube allowance.

  2. 2

    Pick a model

    The default four stem model gives you a vocals stem on every plan. On Studio, choose Karaoke (vocals / instrumental) or Premium vocals when you only want voice and band apart.

  3. 3

    Mute or solo the vocals

    When the track opens, press Mute on the vocals row for the instrumental, or Solo for the acapella. Set an A/B loop around any section you want to rehearse.

  4. 4

    Download your version

    On a paid plan, save the stems as MP3. On Studio, save WAV, a custom mix with your key and levels, or only the A/B region you selected.

Instrumental, acapella or both from one upload

A vocal remover and an acapella extractor do the same job from two directions. The AI estimates which parts of the mix belong to the voice and which belong to everything else, and writes them out as separate stems. You choose afterwards which one to hear, so there is no need to process the song twice.

With the default model the song comes back as four stems: vocals, drums, bass and other. Muting the vocals leaves the full band, but you can keep going: drop the drums as well for a sparse acoustic feel, or solo the bass to hear what it is doing under the chorus. On Studio, the six stem model also gives guitar and piano rows of their own.

For singers: a backing track you can perform with

Taking out the lead vocal is usually the first step, not the last. With the vocals muted, loop the bridge you keep missing, pull the drums back if they crowd your monitor mix, and leave a little of the original singer in for the passages you are still learning.

Studio adds what turns the track into yours. Key control shifts the song so it sits where your voice is comfortable, and the chord lane renames the chords to match. A custom mix then carries your levels, mutes, edits, fades, EQ, key and effects into a single file you can play at a rehearsal or a gig. For a louder, more finished sound, HD Master can render that mix again, practice key and all.

Rehearse the hard lines on a loop

Set A and B around the passage you want to drill, by dragging the markers on the waveform or typing the times, and the player repeats it until you move on. The 10 and 30 second jump buttons get you around the song quickly. The loop is part of every plan, Free included; slowing the song down and the rest of the practice tools are covered on the music practice page.

Keep the vocal, change the singer

The acapella you just isolated is also the input for a voice switch on Studio. From the vocals row’s ⋮ menu, have the stem sung again by a stock AI voice of the other gender, male to female or female to male, with the same words and melody; the melody moves by a whole octave, so your backing track needs no change. The new voice arrives as its own stem and the original is muted, which makes it a quick way to hear a song in a different register before you commit to a key change.

Why a trace of the voice can remain

No vocal remover gets every recording perfectly clean, and it helps to know where leftovers usually hide. Backing vocals and harmonies are sometimes heard as part of the band, so a faint choir can stay in the instrumental. Long reverb and echo tails from the lead can hang on after the dry voice is gone. Vocals that were doubled or spread wide across the stereo field are harder to lift out in one piece, so some of them may stay in the other stem.

When that happens, try another model: the alternatives in the Studio catalog are built differently and leave different traces. Studio can also run denoise or reverb removal on the vocal stem to dry up an acapella. And Mute A–B silences an A/B range in one stem without touching the others, which takes care of a stray phrase; you can restore it later.

Which plan you need

Free takes the vocals out of up to 20 full-length songs daily using the two core models, and you mute, solo, loop and mix the result in the browser; Free tracks last 24 hours, and saving files needs a paid plan.

Starter, $4.99 a month, downloads every stem as MP3, with 200 audio minutes a month and stems kept for as long as you subscribe.

Studio, $9.99 a month, adds the Karaoke and Premium vocals options with the rest of the model catalog, key control, lossless WAV, custom mixes, A/B region downloads, denoise and reverb removal, with 500 audio minutes a month.

Compare every plan

Questions

Can I remove vocals from a song for free?

Yes. On Free you can split up to 20 full-length songs daily, and each one has a vocals stem you can mute for karaoke or solo for an acapella. Listening, mixing and looping all happen in the browser; keeping the files takes a paid plan.

How do I make an acapella from a song?

Separate the song, then press Solo on the vocals stem. Everything else goes quiet and only the voice plays. Download it as an MP3 stem on any paid plan, or as lossless WAV on Studio.

Which model removes vocals best?

Start with the default four stem model, which runs on every plan. If the instrumental still carries too much voice, Studio offers the Karaoke (vocals / instrumental) option and the Premium vocals model, both of which split the song into a vocal and an instrumental only.

Can I change the key of the karaoke track?

Yes, on Studio. Key control transposes the song across 12 semitones, and a custom mix or HD Master exports it in the key you set.

Why can I still hear the singer faintly?

Backing vocals, reverb tails and doubled parts often stay partly in the instrumental. Try a different model, or on Studio run reverb removal or denoise on the stem that needs it.

Can I download only one section of the backing track?

On Studio, yes: set A and B around the section and download just that region, which is handy for a short practice clip or a single verse.

AI stem splitter

Split any song into vocals, drums, bass, guitar, piano and more with AI models, then play, mix and export the stems in your browser. Free plan included.

Music practice tools

Practice with any song: slow it down without changing pitch, change the key, loop a hard section, follow live chords and BPM, and solo the part you play.

Online audio mixer

Mix stems in your browser on one synchronized timeline: volume, mute and solo, silence or fade a section of one stem, arrange the parts and export the exact mix you hear.

AI vocals and lyrics

Sing new lyrics over a track you separated: an AI voice over your own instruments, more takes of the same words and a backing chorus on the sections you pick.

AI voice changer

Switch any vocal stem to a male or female voice with AI: the same words, melody and key, sung by an AI voice of the other gender and added as a new stem beside the original.

Try it on your own track

Start free, no card required