1 TB free audio storage

Pull any instrument out of any song.

Upload a track and ask for the stem you want: drums, bass, vocals, or the rest. Scarleta splits the song into separate sources and hands you clean, downloadable stem files right in the chat.

How it works

No setup, no plugins. Just ask, and get the file back.

Step 1

Upload your song

Drop an audio file into the Scarleta chat: a full mix, a sample, or a track you want to take apart. No plugins to install, nothing to set up.

Step 2

Ask for the stem you want

Type it in plain words: "give me just the drums" or "split this into stems." The AI figures out what to run. No settings, no code.

Step 3

Download the stems

Get separate stem files back: drums, bass, vocals, and the rest, ready to download and drop into your DAW.

Hear the difference

Hear a stem come out of the mix

The full mix before, and one clean isolated stem after.

Stem isolated

Before · Full mix

0:00 / 0:00

Built-in storage

Every result is saved to your library, 1 TB free.

The files you bring and the stem separation results you get are kept in one place, so you can come back, compare earlier versions, share a link, or ask across everything you’ve stored, whenever you want.

1 TB free Share with a link Ask across your collection
Saved to your library1 TB free
Stem separation result
Ready to play, share, or reuse
Play Share Versions

Recipes

Make stem separation one step in a pipeline you run by name.

Chain it with other steps into a recipe you design once, then run on a single file in chat or on thousands at once.

Sample digger
Separate stemsDetect key & tempoTag & file
Instant acapella
Isolate vocalsClean up noiseExport
Remix starter
Separate stemsMatch a referenceEven out loudness

See it in action

See stem splitting in action

Simple, transparent pricing

Only pay for the audio you actually process.

$0.10/ minute

1 token = 1 second of audio, minimum 1 token per job.

Prepaid, no subscription 300 free tokens to start Top up from $10

More than one trick

Scarleta does a whole lot more.

The same account handles all of it. Here are a few very different things you can do with your audio.

For developers

Do this at scale with the API

Running a whole catalog? Send a batch of files to a stem-separation recipe and pass a webhook_url to get one callback when the whole batch is done. Batch submits need a paid account; a free key gets 403 batch_paid_only.

curl https://api.scarleta.ai/v1/batch \
  -H "Authorization: Bearer $SCARLETA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "recipe_id": "rec_1002_cat_vocal_stems",
    "audio_urls": [
      "https://example.com/song.mp3"
    ],
    "webhook_url": "https://your-app.com/hooks/scarleta"
  }'

Split your first song on your free tokens

Upload a track, ask for the stems, and download the separate files. Top up only when you need more.

Frequently asked questions

What stems do I get?

A song is split into a fixed set of sources: drums, bass, vocals, and other, each as its own separate audio file. Every stem comes back as a real file you can download.

Can I get just one instrument?

Yes. Ask for the stem you want ("just the drums", "just the bass") and take that file. The separation produces each source as its own track, so you can keep only the ones you need.

What does splitting a song into stems cost?

Billed at $0.10 per minute of audio, from prepaid tokens (1 token = 1 second): a 4-minute song is 240 tokens, or $0.40. New accounts start with 300 free tokens, top-ups start at $10, and there is no subscription.

Do I need music software to split stems?

No. Upload a track in the Scarleta chat and ask. It runs in your browser, nothing to download.

Can I split stems for many songs from my own app?

Yes. Developers can send many songs at once through our API and get called back by webhook when the batch is done. See the docs. Batch needs a paid account.