---
name: music-hum
description: "From the person's hum, sung or rapped line (a voice message or a recording) to a full song on Suno V6 that keeps their melody and flow — listen (Scribe v2 words + Gemini 3.8 Flash genre/tempo/theme), review words, genre and lyrics, write lyrics around their line, then cover the seed with an external music tool. Part of the music reference set."
---

These creative references describe source experiments with external tools. Use your own supported account and tools; this download installs no automation. Settings and timings are experimental observations, not current FlashBoards availability. Use only media and voices you have permission to use.

# music-hum · a hum or a line → a song

The seed is `{seed}`: the person’s own voice message or audio file. Everything below is computed fresh for it.

## Steps (review one creative choice at a time)

1. **Listen** — transcribe your recording with Scribe v2 and describe its sound with a supported audio-understanding tool such as Gemini.
   Keep the transcript and word times, any alternate reading, whether the words agree, the performance kind, melody, genres, mood, theme and song idea in your project notes. These are observations, not required private JSON keys.
2. **Words** — if the transcriptions disagree or a word is uncertain, listen to that passage and record your intended spelling. If they agree, keep the words; tool attribution is not a reason to prefer one reading.
3. **Vibe** — look up 3 candidate genres from the sound description in MusicBrainz and compare each with "like <A>, <B>, <C>" references. Keep the chosen vibe's style description for the draft. If you already chose a genre or artist, use that choice directly (an artist → `music-refs`).
4. **Lyrics** — decide whether to write new lyrics around the seed, supply your own lyrics, or make only a beat (`instrumental: true`). Skip this decision if already settled.
5. **Draft** — write the lyrics and style by `music-write` (your line is the hook; the style starts from the chosen vibe). Review the full lyrics as they will be sung. Generate only after deciding whether the words or style need changes.
6. **Make** — submit a cover request with your seed and the settings below through a supported external tool. A song took 25 s to 3 min in the source experiments (one run took 121 s). The following JSON is an illustrative authoring plan, not an installed runner or FlashBoards API:
   ```json
   {"name": "{name}", "title": "<title in the seed's language>", "source": "{seed}",
    "style": "<music-write style line>", "lyrics": "<lyrics with [Verse]/[Chorus] tags>",
    "vocal_gender": "<m or f, chosen for the intended singer>", "audio_weight": 0.65, "duration": 150}
   ```
   `audio_weight`: 0.75 when they want it close to the hum, 0.5 when they want it reworked. `duration` 120–180 for a song.
7. **Takes** — listen to both returned tracks and choose:
   keep take 1, keep take 2, or request another round with any intended changes written in your notes. Then `music-after`.

## Pitfalls

- **No lyrics → instrumental.** The source workflow refused a vocal song without `lyrics`; don't work around it with empty tags.
- A seed shorter than ~6 s or pure speech gives a weak melody: say so in a line and offer the idea route (a new song request without `source`).
- `vocal_gender` only raises the chance; if the take comes in the wrong voice, run again rather than changing the style.
- Written corrections and preferences take precedence over a tool's suggestions.

## Example run (reference only — never reuse its words or files)

9.10.2026, the owner's 8.3 s Russian rap line. listen: kind rapping, ~135 BPM, trap / cloud rap / hip hop, Scribe and
Gemini split on one word (3.6–4.0 s) → a words card in the historical prototype. Lyrics: the seed's two lines as the hook, two 8-line verses, a bridge.
make with the seed as `source`, `audio_weight` 0.65, `duration` 150 → two 2:30 takes · 63 s.
Same-day source chat prototype (not an installed or integrated FlashBoards feature): voice message → listen → words card (answered with their own spelling in the free
field) → vibe card (Cloud Rap) → lyrics card (write them) → draft + «Make it?» → two 2:29 takes in 121 s → «Which one?» →
«What next?» (stems + cover art: Suno's picture copied at once, stems landed a minute later). These cards describe that historical prototype, not a reader requirement; no example audio or chat automation is delivered.