# MiniMax Music Cover - Core.Today AI API

> Reimagine any song in a different style. The model extracts the melodic structure from the input audio and regenerates the track — the melody and duration stay the same, but voice, instruments, genre, and arrangement can all change.

- **Provider**: MiniMax
- **Model ID**: minimax/music-cover
- **Category**: Audio & TTS
- **Credits**: 700 per generation (1 audio file)
- **Speed**: Medium
- **Quality**: High

## Features
- Melody and song duration preserved
- Voice, instruments, genre, arrangement all changeable
- Genre transformation: pop→jazz, rock→bossa nova, ballad→EDM
- Optional lyric override
- Production aesthetics control (vinyl texture, modern production)
- Up to 44.1kHz / 256kbps / WAV output

## Use Cases
- Genre-swapped covers of original songs
- Rearranging demos in different styles
- A/B testing arrangements for a track
- Creative remixes for social content
- Localized covers with new lyrics

## API Endpoint

Base URL: https://api.core.today/v1
Create Prediction: POST /predictions
Get Status: GET /predictions/{job_id}

## Authentication
Header: X-API-Key: YOUR_API_KEY

## Input Parameters

### Required
- **prompt**: string - Target style (max 2,000 chars). Name the genre (strongest style shift), vocal style, instruments, and production qualities. e.g. 'Jazz arrangement, saxophone lead, smooth female vocal, mellow piano chords'
- **audio_url**: string - URL of the song to cover. Publicly accessible MP3 or WAV. Works best with songs that have clear vocals and melody

### Optional
- **lyrics**: string - Optional lyrics to override the ones extracted from the source song. If omitted, the original lyrics are sung. Use \n and structure tags like [Verse], [Chorus]
- **sample_rate**: integer (default: 44100) - Sample rate for the generated music
  Options: 16000, 24000, 32000, 44100
- **bitrate**: integer (default: 256000) - Bitrate for the generated music
  Options: 32000, 64000, 128000, 256000
- **audio_format**: string (default: mp3) - Output format. Use wav for production work
  Options: mp3, wav, pcm

## Examples

### Jazz Cover
Turn a song into a smooth late-night jazz arrangement
```json
{
  "model": "minimax/music-cover",
  "input": {
    "audio_url": "https://example.com/original-song.mp3",
    "prompt": "Jazz arrangement, saxophone lead, smooth female vocal, mellow piano chords, late night club",
    "audio_format": "mp3",
    "bitrate": 256000,
    "sample_rate": 44100
  }
}
```

### EDM Remix
Transform a ballad into a driving EDM track
```json
{
  "model": "minimax/music-cover",
  "input": {
    "audio_url": "https://example.com/ballad.mp3",
    "prompt": "EDM remix, 128 BPM, synth bass, atmospheric pads, driving beat"
  }
}
```

### Lo-fi Version
Chill lo-fi rework with vinyl texture
```json
{
  "model": "minimax/music-cover",
  "input": {
    "audio_url": "https://example.com/pop-song.mp3",
    "prompt": "Lo-fi hip hop version, vinyl crackle, chill beat, warm midrange"
  }
}
```

## Response Format
```json
{
  "job_id": "abc123",
  "status": "pending | processing | completed | failed",
  "result": "URL or data (when completed)"
}
```

## Usage Flow
1. POST /predictions with model and input -> receive job_id
2. GET /predictions/{job_id} -> poll until status is completed or failed
3. Result contains output URL(s)

## Tips
- Name the target genre first — it produces the strongest style shift
- Describe the vocal style and lead instruments to shape the new arrangement
- Use a clean full-mix source with clear vocals and melody — heavily compressed sources degrade quality
- Leave lyrics empty to keep the original lyrics; provide lyrics only to replace them
- Melody and duration are preserved; generation takes a few minutes depending on song length
- Only cover songs you have the rights to — respect copyright of the source material

## Documentation
https://replicate.com/minimax/music-cover