# Suno Vocal Separation - Core.Today AI API

> Separate a Suno-generated track into clean vocal and instrumental stems. Takes the taskId and audioId from a previous Suno music generation and returns separated vocalUrl/instrumentalUrl audio files.

- **Provider**: Suno
- **Model ID**: suno/vocal-separation
- **Category**: Audio & TTS
- **Credits**: 120 per separation
- **Speed**: Fast
- **Quality**: High

## Features
- Clean vocal + instrumental stem separation
- Chains directly from Suno music generation output
- Both stems mirrored to CDN as downloadable MP3s
- Karaoke and remix ready

## Use Cases
- Karaoke versions of generated songs
- Acapella extraction for remixes
- Instrumental background versions
- Vocal-only samples for production

## 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
- **taskId**: string - Suno task id from a previous music generation result (the taskId field in the generation output)
- **audioId**: string - Track id to separate — the sunoData[].id field from the generation output

## Examples

### Separate vocals from a generated song
Pass the taskId and one track's id from a suno/music-* generation result
```json
{
  "model": "suno/vocal-separation",
  "input": {
    "taskId": "5c7998a1b2c3d4e5f60718293a4b5c6d",
    "audioId": "34fa3e86-2fc8-4f8a-8ea3-f58b664bde51"
  }
}
```

## 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
- The music generation output contains both inputs: taskId at the top level and audioId as sunoData[].id.
- Each of the two generated tracks can be separated independently (one call per track).
- Only works on tracks generated through the Suno models on this gateway.

## Documentation
https://docs.sunoapi.org/suno-api/separate-vocals-from-music