STABLE AUDIO · PROMPT RECIPES + FREE BUILDER

Stable Audio Prompt Guide

The prompt format Stable Audio actually responds to, worked examples by genre, and a free tool that generates a Stable Audio-ready prompt from any reference track — measured, not guessed.

Anatomy of a Stable Audio prompt

Stable Audio weights tokens toward the start of the prompt. That single fact shapes the whole format: front-load what must be true, push nuance toward the end.

[genre] [BPM] [energy / mood], [instrumentation], [production], [structure] · [exclusions]

Layer by layer

Stable Audio prompt examples (copy-paste)

Lo-fi hip hop · 84 BPM

lo-fi hip hop, 84 BPM, lazy and nostalgic. Dusty drum machine with swung shuffled hats, upright bass, warm Rhodes chords, vinyl crackle texture. Tape-saturated master, narrow brights, laid-back half-time feel. No vocals.

Melodic techno · 126 BPM

melodic techno, 126 BPM, hypnotic and driving. Punchy four-on-the-floor kick, taxi bass, arpeggiated analog sequence slowly opening, cinematic wide pads on top. Clean modern production, deep low end, subtle sidechain to the kick. Builds in 8-bar steps, breakdown at two thirds. Instrumental, no vocals.

Cinematic trailer · 140 BPM

epic cinematic trailer music, 140 BPM, ominous rising tension. Sub-bass pulses, staccato string ostinato, brass swells, granular risers and metallic impacts. Big reverberant hall production, dynamic contrast between sparse stages and full orchestra hits. No melody-carrying synth leads.

Ambient · 72 BPM

ambient soundscape, 72 BPM feel, weightless and introspective. Slow evolving granular pads detuned over stacked octaves, distant piano notes with long reverb tails, soft noise bed. Extremely wide stereo, gentle low-pass movement across the piece. No drums, no vocals.

More ready-made prompts across moods and engines: AI music prompt examples.

Prompt length and what to leave out

Generate a Stable Audio prompt from a reference track

The hardest part of a Stable Audio prompt is the honest numbers: tempo, rhythm density, brightness, dynamics, stereo width. Music DNA Studio measures them from any MP3, WAV, FLAC or MP4 — locally in your browser, nothing uploaded — then outputs a Stable Audio-shaped prompt (and versions for Eleven Music, Soundraw, Mubert, Beatoven and AIVA) with exclusions set. You keep the creative layer; the spec underneath is measured.

Try It On A Track You Love

Drop it in, pick Stable Audio, copy the prompt. Free, no signup, nothing leaves your device.

ANALYSE YOUR AUDIO — FREE →

FAQ

What is the best format for a Stable Audio prompt? ▾

Front-loaded description: primary genre and BPM first, then energy and instrumentation, then production character. Tokens early in the prompt carry more weight.

Does Stable Audio respect BPM in prompts? ▾

Yes. Explicit tempo like "126 BPM" is one of the tokens Stable Audio honors most reliably, so put it right after the genre.

How do I get an instrumental track from Stable Audio? ▾

State it explicitly — "instrumental, no vocals" — near the front of the prompt, and again in the negative/exclusion field where the version you use supports it.

Can I write a Stable Audio prompt based on a song I like? ▾

Yes — describe the song's measurable properties (BPM, rhythm density, brightness, dynamics, mood, instruments) instead of naming the artist. Music DNA Studio analyzes any track in-browser and generates a Stable Audio-formatted prompt automatically.

Why does my prompt keep producing the same generic result? ▾

Usually a missing tempo, a missing production layer, or contradictory genre tags. Change one layer at a time — rerolling an unchanged prompt tends to produce nearly the same output.