Make the Instrumental of Any Song
Pull the vocal and keep the full band: a clean instrumental you can record a cover over, film over, sample or perform with. Tune the blend in a live mixer, then download MP3 free, or a WAV bounce with Pro when it’s going somewhere that matters.
Short answer: to make an instrumental of a song, upload the audio file to StemConsole and run Vocal Remover. The AI separates the vocal from the music, you check the instrumental in the mixer, and download it as MP3 for free, or as WAV with Pro. Your first split needs no account.
How it works
Upload the song
The track you want the instrumental of: MP3, WAV, FLAC, OGG, M4A, AAC or WMA, up to 200MB.
Vocals come out
The AI removes the lead vocal and leaves the instrumental intact.
Mix & download
Audition it, blend stems to taste, then download MP3 (Free) or WAV (Pro).
An instrumental you can actually use
An instrumental is only useful if it holds up when it is the main thing people hear. StemConsole opens your split in a live mixing console in the browser: the instrumental and the vocal each get a channel strip with solo, mute and level, real waveforms and metering. Audition it before you download, blend to taste, and bounce exactly the version you mixed.
Want a hint of the original vocal left in as a guide for a singer, or the drums pushed up for a dance video? Move a fader instead of settling for all or nothing. Then download MP3 on Free, or a WAV bounce with Pro for recording and release work. No watermarks, ever.
I’m Dan Murtagh, a mixing and mastering engineer with 20 years of records behind me and a decade teaching audio in Melbourne. What follows is the advice I give students who want a backing track that sounds finished, not like a leftover.
- Instrumental plus the isolated vocal from every split
- Quick Split (free) for vocals, drums, bass and other when you need more control
- MP3 downloads and bounces free, WAV stems and WAV bounces with Pro
- Saved sessions in My songs and share links
- Runs in the browser on phone and desktop, separation on GPUs
What makes a clean instrumental
The separation is only as good as what goes in. These are the factors I listen for when a student brings me an instrumental that “doesn’t sound right”.
Reverb and delay bleed
The dry vocal separates cleanly. Long reverbs and delay throws are spread across the stereo image and blend with pads and guitars, so a soft wash of them can remain in the instrumental, most noticeably in the gaps between lines. Under a new vocal it usually disappears; in a quiet breakdown under a video you may notice it.
Backing vocals and doubles
Harmonies, doubles and ad-libs go into the vocal stem along with the lead. For a cover, that is normally right, because you will sing your own. If you want the harmonies in your backing, bring the vocal fader up a touch and bounce that blend.
Modern masters vs old recordings
Modern productions, close-miked and heavily compressed, tend to give very clean instrumentals because the vocal is distinct from the band. Vintage recordings are more blended: mono or narrow stereo, singer and band in one room, tape noise. They work, with more ghosting of the vocal to listen for.
Low-bitrate rips
A low-bitrate MP3 or streaming rip has already lost detail and gained warbly artefacts around the voice. Those artefacts get split across both stems and can make the instrumental sound watery. If the instrumental is going into a release or a video, find a lossless or high-bitrate source first. WAV vs MP3 for stems goes deeper.
How to check it before you use it
Skip to the loudest, busiest chorus, solo the instrumental and listen on headphones at a moderate level. Listen for a ghost vocal: a thin, phasey trace of the melody, or S and T sounds poking through. Then listen to the quietest section, where reverb tails are most exposed. If both pass, the whole track will.
Practical fixes
- Re-split from a better file. This fixes more problems than any processing.
- Leave a little vocal in on purpose. For rehearsal or a guide track, a low vocal level masks leftovers and helps performers stay on the form.
- EQ the instrumental. A gentle, wide dip somewhere around 2 to 5kHz clears space for a new vocal and hides residue.
- Use more stems. Quick Split lets you rebalance drums, bass and other, and Full Separation adds guitar and piano, so you can fix an arrangement that feels thin once the vocal is gone.
Ways to get an instrumental, compared
| Method | What you get | Trade-off | Best for |
|---|---|---|---|
| AI separation (StemConsole) | Instrumental and vocal stems from a normal stereo file, plus 4 or 6 stems if you need them | Quality depends on the source file and how the vocal was mixed | Covers, rehearsal, video beds, sampling, most songs |
| Phase cancellation (“center channel” removal, the Audacity invert trick) | Everything not panned dead centre | Removes centre bass, kick and snare too, leaves stereo reverb, does nothing on mono files | Quick tests on simply panned older mixes |
| Official instrumentals or multitracks | The label’s own instrumental or the session stems | Only exists for some songs, and releasing still needs a licence | Sync, releases and licensed projects |
Always check whether an official instrumental exists: for singles it often does. When it doesn’t, AI separation is the practical option, because phase cancellation hollows out the low end that makes a backing track feel full. The full walkthrough is in how to remove vocals from a song.
What to do with your instrumental
Covers for YouTube and TikTok
Record your vocal over the real arrangement so the cover sounds like the song, not a MIDI version of it. Use WAV with Pro so the backing is not compressed twice, once by you and once by the platform. If the original key does not suit your voice, Pro can change the key without changing the tempo. Before you post, read the rights question in the FAQ below.
Video beds and content
A vocal-free version sits under narration far better than the full song, because nothing fights your voice in the midrange. Bounce the instrumental at a lower level, or mute the drums with Quick Split for a softer bed under dialogue.
Sampling and beat-making
Grab a clean section of the instrumental to chop, flip or rap over, then loop the section in the mixer to find the right bars. If you want the vocal for a remix instead, the acapella extractor is the page for that, and acapella vs instrumental explains the difference in use.
Auditions and live performances
Singers preparing for auditions, school productions or a gig can rehearse with the band arrangement they will hear on the night. Pro’s count-in and speed change without pitch change help you learn tricky sections before running the whole thing at tempo. For a karaoke-style version with a guide vocal, see how to make a karaoke track.
Worship teams and backing tracks
Worship and function bands often need to fill a missing part: no keys player this week, or a small team covering a big arrangement. Split the song, mute the parts your musicians will play, and bounce the rest as a backing track with the backing track maker. Full Separation gives you guitar and piano as separate stems for exactly this.
Free or Pro?
Free gives you the instrumental maker in full: Vocal Remover and Quick Split, 5 splits a month with up to 2 Quick Splits, MP3 downloads and bounces, the live mixer with basic looping, saved sessions and share links, with no watermarks. Pro is $6.99 a month or $49 a year (USD), with a 7-day free trial for eligible accounts. It adds WAV stems and WAV bounces, Full Separation into 6 stems, 100 splits a month, speed and key change, named loops, count-in, tap tempo and priority processing. For a single song that needs 6 stems, a $1.99 credit unlocks one Full Separation and never expires.
Frequently asked questions
How do I make an instrumental version of a song?
Upload the song to StemConsole and choose Vocal Remover. The AI takes the vocal out and leaves the instrumental, which you can check and adjust in the live mixer, then download as MP3 free or WAV with Pro. There are no watermarks and nothing to install.
Can I make an instrumental of any song?
Of almost any recording you have as an audio file. Modern, well-produced songs in a good-quality file give the cleanest instrumentals. Vocals with heavy reverb, stacked backing vocals, old mono recordings and low-bitrate rips leave more trace behind, so check the biggest chorus before you commit.
Why can I still hear faint vocals in my instrumental?
Usually it is reverb or delay from the vocal, backing vocals, or compression artefacts in a low-bitrate file. Re-split from the best-quality copy you have. Once you record a new vocal or add a voiceover on top, most of that trace is masked, and a gentle EQ dip around 2 to 5kHz on the instrumental helps too.
Can I upload a cover made with the instrumental to YouTube or TikTok?
Making an instrumental for personal practice, rehearsal or study is how most people use it. Publishing a cover over the original recording uses the master recording as well as the song, so you need permission or a licence from the rights holders, and platforms may claim, mute or block it. StemConsole does not grant those rights.
Is the instrumental full quality?
Free downloads are MP3 with no watermark, which is fine for practice, rehearsal and rough demos. For a cover recording, a video edit or anything you will release, Pro gives you WAV downloads and a WAV bounce of your exact blend.
Can I keep the backing vocals in the instrumental?
The vocal stem includes the lead and most backing vocals, so they come out together. If you want some voice back, raise the vocal fader slightly in the mixer and bounce that blend. Solo each stem first to hear exactly what is in it.
What’s the difference between the instrumental maker and the vocal remover?
They run the same 2-stem split and both give you the instrumental and the vocal. The vocal remover is the general page. This page focuses on using the instrumental as a finished backing track for covers, videos, sampling and performances. For singing along, the karaoke maker is set up for that.
Does it work with YouTube links?
No. StemConsole does not import YouTube or streaming links. Upload an audio file you have: MP3, WAV, FLAC, OGG, M4A, AAC or WMA, up to 200MB. A lossless or high-bitrate file gives a noticeably cleaner instrumental than a rip.
How long does it take?
After the upload, separation runs on GPUs rather than your device, and most songs are ready in the mixer within a couple of minutes. Pro splits get priority processing.
Can I get just the drums and bass, without guitar or keys?
Yes. Quick Split (free) gives you vocals, drums, bass and other, so you can mute more than the vocal. Full Separation with Pro, or a $1.99 credit for one song, adds separate guitar and piano stems. Mix what you want and bounce it with the backing track maker.
More tools and guides
Vocal Remover
Strip the vocal from any track: clean instrumental plus isolated vocal.
Karaoke Maker
Turn any song into a karaoke-ready instrumental in minutes.
Make a backing track
Set each stem’s level, then bounce the mix: MP3 free, WAV with Pro.
Acapella Extractor
Pull clean acapellas for remixing and sampling.
Change a song’s key
Move the instrumental into your vocal range without changing speed.
StemConsole vs StemSplit
Per-song vs per-minute pricing, free tiers and the live mixer.