How Soundraw actually takes direction
Soundraw is parameter-driven, not free-text. You do not describe a track in a sentence — you pick a genre, two or three moods, a tempo band and a length, and Soundraw generates inside that box. Which means the "prompt" is the combination, and the combination has to agree with itself.
Setting by setting
- One primary genre. "lo-fi hip hop" — not "lo-fi hip hop / jazz / ambient". Extra genres do not blend; they blur the selection space.
- Moods that agree. Soundraw lets you stack moods, but they still have to point the same direction. "Nostalgic + relaxed + warm" holds together; "dark + uplifting" pulls the result toward neither. Two or three is the sweet spot.
- Tempo band from the real number. Soundraw asks for slow / medium / fast. If your reference track measures 84 BPM, that is the slow band — do not guess "medium" because it feels moderate.
- Length that matches the use. Set it to where the content actually ends. A 90-second social edit and a five-minute background bed want different arcs.
- Structure edits after generation. The section timeline is where "generic" becomes "yours": fade the intro, lift the middle third, decide how the outro decays.
- Instrument mix. Mute and solo tracks to make the layers you care about prominent — the same job the instrumentation order does in a text prompt.
Soundraw settings examples (copy-paste)
Each example is a complete recipe: the picker selections, then the timeline pass. Copy the settings into Soundraw, then apply the structure edits section by section.
Lo-fi study bed · slow · 84 BPM feel
Uplifting corporate · medium · 116 BPM feel
Dark documentary tension · slow · 76 BPM feel
Beach lounge groove · medium · 104 BPM feel
More ready-made prompts across moods and engines: AI music prompt examples.
What to pick, and what to leave out
- Leave out conflicting moods. The most common reason a Soundraw track lands generic is a mood pair that cancels out. Pick a lane.
- Leave out artist names. They describe less than mood and tempo words do, and they trigger content filters.
- Leave out the wrong tempo band. A "fast" band with "relaxing" moods fights itself; match the band to the energy the moods imply.
- Do the timeline pass. Regenerating without structure edits just gives you another take of the same idea — section edits are the creative layer.
Build a Soundraw setup from a reference track
The hard part of a Soundraw recipe is the honest numbers: tempo band, mood words that match the actual energy, brightness, dynamics, instrumentation. Music DNA Studio measures them from any MP3, WAV, FLAC or MP4 — locally in your browser, nothing uploaded — then outputs a Soundraw-shaped settings sheet (and versions for Eleven Music, Stable Audio, Mubert, Beatoven and AIVA) with the exclusions already set. You make the creative calls; the numbers underneath are measured.
Dial In The Recipe
Drop in a track, pick Soundraw, copy the settings sheet. Free, no signup, nothing leaves your device.
ANALYSE YOUR AUDIO — FREE →FAQ
How do I direct Soundraw for the sound I want? ▾
Pick one primary genre, two or three moods that agree with each other, a tempo band, and a length. Treat that combination as the prompt — Soundraw generates inside it.
Does Soundraw accept text prompts? ▾
Not as its primary input — it generates from genre, mood, tempo and length selections, then you refine sections on a timeline. A written description is still useful: it tells you exactly which settings to pick.
Can Soundraw generate vocals or singing? ▾
No — Soundraw output is instrumental. Think background score, underscore and bed music rather than a song with a singer.
Can I match a song I like using Soundraw? ▾
Yes, if you translate the song into measurable properties — tempo, mood words, brightness, dynamics, instrumentation — instead of its name. Music DNA Studio analyzes any track in-browser and outputs a Soundraw-ready settings sheet automatically.
Why does my Soundraw track sound generic? ▾
Usually conflicting mood selections, a tempo band that fights the energy you want, or skipping the timeline structure edits. Fix the mood pair first, then use the section editor — that is where "generic" becomes "yours".