How to Add Instruments to a Song

Layer is Song Creator Pro's tool for adding accompaniment to an existing track. You upload a single-instrument recording (a vocal, a guitar, a piano melody) and Layer generates the rest of the band around it, keeping your original audio as the anchor.

It's ideal for turning rough demos into fuller arrangements, adding a backing band to an a cappella recording, or fleshing out a melody idea.

This guide walks you through using Layer.

When to Use Layer

Layer is the right tool when you want to:

• Turn an a cappella vocal into a full song: add drums, bass, and instruments around your voice
• Build out a guitar or piano sketch into a complete arrangement
• Flesh out a demo with an AI-generated band
• Add accompaniment to a vocal stem you extracted from another track (see the Extract guide)
• Experiment with different genre arrangements of the same vocal: try the same source as pop, then rock, then lo-fi, then orchestral

If you want to replace your source with a new version, use Remix instead. If you want to change only a portion of an existing full track, use Revise.

Step 1: Upload Source Audio

Click the Layer tab at the top of the interface. In the Source Audio field on the left, upload the track you want to build around.

Layer works best with single-instrument or mostly-single-instrument sources:

• A vocal recording (dry or with light reverb)
• A solo instrument: acoustic guitar, piano, synth lead
• A vocal stem extracted from another song (use Extract to pull one out)
• A humming or melodic idea recorded on your phone

Cleaner source audio equals cleaner results. If your source is a noisy voice memo, expect the generated accompaniment to have more character but less polish.

Step 2: Choose Instrument Hints (Or Leave on Auto)

At the top of the left panel, you'll see a grid of Instrument pills: Vocals, Backing Vocals, Drums, Bass, Guitar, Keys, Perc, Strings, Synth, Brass, Winds, FX.

By default these are all off, and Layer runs in Auto mode ("model decides what to add"). The model will pick an instrumentation that fits your description and source audio. For most starting points, Auto is the right choice.

Click any pill to toggle it on. Selected pills act as hints telling the model which instruments you want in the accompaniment:

• Want a stripped-back arrangement? Pick only Drums and Bass.
• Want a full rock band? Pick Drums, Bass, Guitar, Keys.
• Want lush orchestration? Pick Strings, Brass, Winds, Perc.
• Want an electronic bed? Pick Synth, Perc, FX.

Note: Instrument hints are suggestions, not guarantees. The model prioritizes the instruments you pick but may still add others if they make musical sense in context.

Step 3: Write a Description

The Description field defines the style of the accompaniment Layer generates. This is where you steer the direction.

Be specific about genre, mood, and production style. Let the Instrument pills handle "what plays" while the description handles "how it sounds." See the prompting guide for more tips on writing effective descriptions.

Example descriptions:

For a vocal → acoustic singer-songwriter:

acoustic singer-songwriter, fingerpicked guitar, light percussion, warm intimate production, mellow and reflective

For a piano → cinematic:

cinematic orchestral accompaniment, swelling strings, subtle brass, dramatic build, grand and emotional

For a vocal → lo-fi hip-hop:

lo-fi hip-hop beat, dusty drum samples, mellow rhodes, vinyl crackle, relaxed bedroom production

Step 4: Set (or Keep) the Lyrics

If you want vocals in the result, add your lyrics to the Lyrics field, broken into [Verse], [Chorus] sections. The model uses these to shape the accompaniment around the song structure.

If you want a purely instrumental result, leave the Lyrics field empty and add "instrumental" to your description.

Step 5: Music Details

The BPM, Duration, Key, Time Sig, Language fields in the Music Details panel usually auto-fill from your source audio. Leave them unless you specifically want to override.

If your source has an obvious tempo or key and the auto-detect gets it wrong, set it manually. Layer is sensitive to BPM and key mismatch between the source and the generated accompaniment.

Step 6: Generate

Hit Generate in the top right.

Batch generate at least 2–4 versions. Layer is especially worth iterating on. The same vocal source with the same description can produce very different arrangements, and often the second or third take is noticeably better than the first.

Each result keeps your original source audio as the foundation and wraps new accompaniment around it. Listen to all of them, then use the Save button on your favorite to export it.

Creative Workflow Ideas

Turn any recording into a finished song. Sing a melody into your phone's voice memo app. Upload it as Source Audio in Layer. Add a description like "indie pop band, driving drums, catchy guitar hook, upbeat." Generate. You now have a finished-sounding track wrapped around your idea.

Genre-surf the same vocal. Upload a vocal source. Generate one result as "lo-fi hip-hop." Change the description to "orchestral cinematic" and generate again. Then "80s synthwave." You'll end up with the same vocal in five different universes.

Extract → Layer. Use Extract on a song you like to isolate its vocal. Upload that vocal stem to Layer and generate a new arrangement around it. Instant genre-bent cover.

Demo to arrangement. Record a rough guitar-and-vocal demo. Layer it with a description matching your target genre. Use the result as your arrangement guide for a real recording, or just keep the AI version.

Related Guides

• Remix: reinterpret a whole track in a new style
• Revise: regenerate a specific part of a track
• Extract: pull individual stems out of a song
• Getting Started Guide: the basics of generating songs from scratch

---

Ready to add instruments to your songs? Get Song Creator Pro on the Microsoft Store for $49.99. One-time purchase, unlimited AI music generation, runs entirely on your PC. Also available on itch.io for $44.99.