# Best Prompts for AI Music

Writing good prompts is the single most important skill in AI music generation. A great prompt produces a great song. A vague prompt produces forgettable noise. This guide covers the techniques that consistently produce high-quality results in MusicWave.

## The 5-element prompt formula

Every strong prompt includes these five elements:

1. **Genre** — the broad musical style
2. **Mood** — the emotional tone
3. **Instruments** — what plays
4. **Tempo** — the speed
5. **Context** — what the music is for

A prompt missing any of these tends to produce inconsistent results.

### Example using the formula

> \[Genre: indie folk] \[Mood: hopeful and warm] \[Instruments: acoustic guitar, light piano, soft drums] \[Tempo: 90 BPM] \[Context: background music for a heartfelt vlog]

This prompt gives the AI everything it needs to produce a focused result.

## Use specific genre names

"Pop" is too broad. The AI generates random pop. Specific subgenre names produce focused results.

| Vague      | Specific                     |
| ---------- | ---------------------------- |
| Pop        | 80s synth pop                |
| Rock       | 90s grunge rock              |
| Electronic | Tropical house               |
| Hip-hop    | Lo-fi boom bap               |
| Classical  | Baroque chamber music        |
| Jazz       | Cool jazz with brushed drums |

The more specific the genre, the more the AI knows what to produce.

## Reference real artists or songs (carefully)

You can reference styles to anchor the AI:

* "In the style of early Coldplay"
* "Similar to lo-fi hip hop beats"
* "With energy like 80s synthwave"

Avoid asking for exact copies of copyrighted music. Use style references to guide direction, not to clone.

## Specify instrumentation clearly

List the instruments you want. Be specific:

* Instead of "guitar" → "fingerpicked acoustic guitar"
* Instead of "drums" → "brushed jazz drums" or "punchy electronic kick drum"
* Instead of "piano" → "warm grand piano" or "bright upright piano"

Limit to 3-5 main instruments for clean results. More instruments often produces muddy mixes.

## Use tempo descriptors or BPM

Tempo dramatically changes the feel. You can specify:

* **General terms:** slow, medium, fast, very fast
* **BPM ranges:** "around 120 BPM"
* **Reference:** "danceable tempo"

Common BPM ranges by genre:

| Genre       | Typical BPM |
| ----------- | ----------- |
| Ballads     | 60-80       |
| Hip-hop     | 80-100      |
| Pop         | 100-130     |
| House       | 120-130     |
| Techno      | 130-145     |
| Drum & Bass | 160-180     |

## Describe mood with emotion words

Mood words help the AI choose chord progressions, melodies, and dynamics. Effective mood words:

* **Positive:** uplifting, joyful, hopeful, energetic, triumphant
* **Negative:** melancholic, somber, dark, anxious, mournful
* **Neutral:** contemplative, focused, mysterious, dreamy
* **Intense:** aggressive, powerful, dramatic, urgent
* **Calm:** peaceful, serene, gentle, tranquil

Avoid vague words like "cool," "nice," or "good."

## Describe the song's structure

You can guide the arrangement with structural hints:

* "Slow build with drums entering at 0:30"
* "Quiet verse, loud chorus dynamic"
* "Sparse intro, full arrangement by 0:45"
* "Climaxes at 1:30 then fades out"
* "Repeats main motif throughout"

Structural cues create more dynamic and satisfying songs.

## Use negative prompting

Tell the AI what NOT to include:

* "No vocals"
* "No drums"
* "No guitar"
* "No electronic elements"
* "No saxophone solo"

Negative prompts work especially well when you want a specific subset of instruments.

## Vocals: be descriptive

If you want vocals, describe them:

* **Gender:** male, female
* **Range:** soprano, alto, tenor, bass
* **Style:** smooth, raspy, breathy, powerful, whispered
* **Emotion:** longing, joyful, angry, peaceful

Example: "Smooth female alto vocals with longing emotion, sparse delivery"

## Examples by use case

### Background music for video

> Soft ambient lo-fi instrumental, no vocals, gentle piano with vinyl crackle, 70 BPM, calm and contemplative mood, loops smoothly

### Workout music

> High-energy electronic dance track at 140 BPM, driving four-on-the-floor beat, big synth lead, motivating and intense mood, builds energy throughout

### Cinematic scene

> Cinematic orchestral piece in D minor, starts with solo piano at 0:00, strings enter at 0:30, full orchestra by 1:00, dramatic climax with brass at 1:45, melancholic but hopeful mood

### Pop song

> Catchy pop song in major key at 120 BPM, female vocals with reverb, electric guitar and synth pads, drums kick in at the chorus, summer love theme, upbeat and nostalgic

### Lo-fi hip hop

> Lo-fi hip hop instrumental, jazz piano sample, brushed drums, vinyl crackle texture, 75 BPM, mellow and chill mood, no vocals, perfect for studying

## Iteration strategy

Rarely is the first generation perfect. Use this workflow:

1. **First prompt** — get a baseline
2. **Listen carefully** — note what's wrong
3. **Adjust 1-2 things** — don't change everything
4. **Regenerate** — see if the change worked
5. **Repeat** — usually 3-5 iterations get to a great result

Don't rewrite the entire prompt every time. Small tweaks reveal what's working.

## Save prompts that work

When you find a prompt that produces consistently good results, save it. Build a personal library:

* "My lo-fi study music prompt"
* "My epic cinematic prompt"
* "My pop song prompt"

You can reuse these as starting points for future generations.

## Common prompt mistakes

1. **Too vague** — "Make a song"
2. **Too detailed** — A 200-word prompt confuses the AI
3. **Contradictions** — "Aggressive calm music"
4. **Too many instruments** — "Piano, guitar, drums, bass, synth, strings, brass, woodwinds, percussion"
5. **Vague moods** — "Cool" "nice" "good"
6. **Missing genre** — Without genre, results are unpredictable
7. **Asking for copies** — "Make Bohemian Rhapsody" — won't work and risky

## Next steps

* [Genre-Specific Prompting](/docs/ai-music-generation/genre-specific-prompting.md) — Templates by genre
* [Generate Your First AI Song](/docs/ai-music-generation/generate-first-song.md) — Practice the basics
* [Lyrics Generator](/docs/tools-documentation/lyrics-generator.md) — Pair with AI lyrics

[Try MusicWave free →](https://www.musicwave.ai)


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://musicwave.gitbook.io/docs/ai-music-generation/best-prompts.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.