---
title: "AI Music Generator & Song Maker for Songs, Lyrics, BGM and Soundtracks"
summary: "Turn a theme, lyric, scene, or reference audio into reviewable songs, instrumentals, BGM, and soundtracks with a clear creative direction and focused revisions."
url: "https://beatra.ai/en/skills/beatra-ai-music-creator"
language: "en"
version: "1.4.7"
last_verified: "2026-11-09"
---

[Skills](https://www.beatra.cn/en/skills) · [Music & audio](https://www.beatra.cn/en/skills/categories/music-audio)

# Turn themes, lyrics, scenes, and references into music

AI Music Creator turns a theme, lyrics, a scene, or reference audio into original songs, instrumentals, BGM, video soundtracks, jingles, and multilingual music as MP3 audio. Each track follows the genre, mood, structure, singable lyrics, vocal direction, and arrangement you set, and a FLAC, MP3, or WAV reference of up to 480 seconds can guide a new arrangement. AI Music Creator works in Claude Code, Codex, Cursor, and other agents that run Skills.

```text
Install AI Music Creator and tell me when I need to sign in: https://beatra.ai/skills/beatra-ai-music-creator/install.md
```

Send it to WorkBuddy, Codex, Claude Code, or another agent to install. Sign in once the first time you use it.

[install.md](https://www.beatra.cn/skills/beatra-ai-music-creator/install.md) · [GitHub](https://github.com/beatra-ai/beatra-skills/tree/main/skills/beatra-ai-music-creator) · [Docs](https://docs.beatra.cn/en/docs/capabilities) · [Use it on your phone: WeChat mini program](https://www.beatra.cn/en/mini-program?utm_source=website&utm_medium=skill)

## Complete music directions and audio you can review

### Develop the whole musical idea

Turn a fragment into a coherent title, hook, lyric, structure, vocal direction, and arrangement.

### Choose the route that fits the destination

Make a vocal song, instrumental, BGM, jingle, soundtrack, multilingual track, or reference-led arrangement from an existing recording.

### Refine the strongest result

Hear every returned version, preserve what works, and turn the biggest gap into one focused next direction.

## How it is made

### Shape the direction before generation · Give every track one coherent musical point of view

Clarify the destination, listener, emotional arc, structure, lyric route, vocal character, instrumentation, and ending before creating audio.

- Start from the real destination and audience
- Write or refine complete singable lyrics when needed
- Record the model, exclusions, reference intent, and ending direction

### Create one clear direction · Hear every version returned from the chosen music model

Use Suno 6 by default or another model you explicitly choose, then keep every returned clip attached to the same creative direction.

- Choose vocal or instrumental before generation
- Use only controls supported by the selected model
- Keep all returned versions available for comparison

### Listen and refine with focus · Preserve the strongest qualities and change the biggest gap

Review composition, lyrics, vocals, pronunciation, arrangement, ending, and production only from audio you can actually hear.

- Compare every returned version
- Separate factual file details from listening judgments
- Make each new direction a deliberate revision

## Songs, BGM, soundtracks, jingles, and multilingual releases

### Original songs from an idea

Develop a story, mood, lyric fragment, or hook into a structured vocal track with a clear musical identity.

### Lyrics into a song

Use complete lyrics you provide or develop singable words before choosing the genre, voice, and arrangement.

### Video, campaign, and brand music

Create BGM, soundtracks, jingles, and campaign music around pace, emotion, dialogue space, and ending needs.

### Games, podcasts, and product experiences

Build themes, intros, transitions, atmosphere, and instrumental cues from a functional brief.

### Multilingual and reference-led directions

Plan bilingual songs or new arrangements from reference audio, then review melody, vocal character, and arrangement against the intended direction.

## Inputs and outputs

| Input or output | Type | Formats | Specifications |
| --- | --- | --- | --- |
| Input | Text — Theme, lyrics, scene, or creative brief | — | — |
| Input | Audio — Reference audio for a new arrangement (optional) | FLAC, MP3, WAV | 6–480 s |
| Output | Audio — Songs, instrumentals, BGM, or multilingual tracks | MP3 | — |
| Output | Text — Singable lyrics | — | editable |

Specifications last checked 2026-11-09 · Sources: [Model catalog](https://docs.beatra.cn/en/docs/models) · [Documentation](https://docs.beatra.cn/en/docs/mcp-tools) · [Terms of Service](https://www.beatra.cn/en/terms)

## Questions

### What can this AI music creator make?

It can develop and create original songs, instrumentals, BGM, video and game soundtracks, jingles, brand music, podcast themes, multilingual tracks, and reference-led new arrangements.

### Can it write lyrics or use lyrics I already have?

Yes. It can develop singable lyrics from a brief or use complete lyrics you provide, keeping the words aligned with the genre, mood, structure, vocal, and arrangement direction you set.

### How can I shape instrumentals, loopable BGM, and track length?

Set the target duration, ending, loop point, energy curve, dialogue space, and instrumental palette, and the track is made to those settings.

### Which music model does it use?

Suno 6 is the creative default. You can explicitly choose another currently available music model when its capabilities fit the brief.

### Can it use reference audio for a new arrangement?

Yes. Say what should carry over from the reference and what should change; the new arrangement follows that direction in melody, vocal character, energy, instrumentation, and arrangement.

## Related Skills

- [Brand Jingle Studio](https://www.beatra.cn/en/skills/brand-jingle-studio): Write the hook, then hear original ad songs and audio-logo stings in the lengths the campaign needs.
- [Exam Mnemonic Song](https://www.beatra.cn/en/skills/exam-mnemonic-song): Write a mnemonic song from the knowledge point you already have, then grow it into a syllabus set.
- [Chinese Poetry Song Composer](https://www.beatra.cn/en/skills/guofeng-poetry-composer): Get the poem's text, readings and setting plan first, then a guofeng song or a guqin-style instrumental with ink-wash visuals.
- [Indie Game Soundtrack and Asset Pack](https://www.beatra.cn/en/skills/indie-game-asset-producer): Get the track list and every line written first, then the soundtrack, the UI and character voices, and the hero art made from them.

[More Music & audio Skills](https://www.beatra.cn/en/skills/categories/music-audio)
