Skip to content

Split Audio Stems#

Split mixed audio tracks into isolated stems for vocals, drums, bass, and other instruments. This tool uses advanced source separation technology to cleanly separate different elements of a song.

Overview#

  • Isolates vocals from instrumentals
  • Separates drums and bass tracks
  • Extracts other musical elements
  • Creates high-quality isolated stems

Quick Start#

  1. Visit Eden Create: Split Audio Stems tool
  2. Upload your audio file
  3. Click "Create"

See API Reference section below for API usage examples.

Ask an Eden agent to "Split Audio Stems".

Basic Usage#

Required input:

  • Audio file (MP3, WAV, or M4A format)

Output stems:

  • Vocals
  • Drums
  • Bass
  • Other instruments
  • Instrumental (no vocals)

Technical Guidelines#

Supported Formats#

  • Input: MP3, WAV, M4A
  • Output: WAV format
  • Maximum duration: 10 minutes
  • Recommended sample rate: 44.1kHz
  • Bit depth: 16-bit or 24-bit

Best Practices for Inputs#

  • Use high-quality source audio
  • Ensure clean, uncompressed audio when possible
  • Avoid heavily distorted or low-quality recordings
  • Keep audio length under 10 minutes

Technical Details#

Base Technology#

  • Model: Librosa
  • Processing: Source separation using deep learning
  • Output format: WAV files

Default Parameters#

  • Sample rate: 44.1kHz
  • Bit depth: 16-bit
  • Channels: Stereo

Limitations#

  • Maximum file size: 100MB
  • Maximum duration: 10 minutes
  • Processing time varies with length

Common Issues and Solutions#

Issue Solution
Poor separation quality Use higher quality input audio
Processing fails Ensure file is under size/length limits
Artifacts in output Check input audio for clipping/distortion
Long processing time Reduce audio length or file size

API Reference#

Endpoint#

POST https://api.eden.art/v2/tasks/create

Headers#

Content-Type: application/json
X-Api-Key: YOUR_API_KEY

Basic Request#

curl -X POST "https://api.eden.art/v2/tasks/create" \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: YOUR_API_KEY" \
  -d '{
    "tool": "audio_split_stems",
    "args": {
      "audio": "https://edenartlab-lfs.s3.us-east-1.amazonaws.com/comfyui/models2/assets/promp+it+up.mp3"
    }
  }'

Response Format#

{
  "task": {
    "_id": "task_12345",
    "createdAt": "2024-01-29T21:52:07.171578+00:00",
    "updatedAt": "2024-01-29T21:52:07.171608+00:00",
    "user": "user_id",
    "requester": "requester_id",
    "tool": "audio_split_stems",
    "parent_tool": null,
    "output_type": "audio",
    "args": {
      "audio": "https://edenartlab-lfs.s3.us-east-1.amazonaws.com/comfyui/models2/assets/promp+it+up.mp3"
    },
    "mock": false,
    "cost": 1,
    "handler_id": "fc-xxxxx",
    "status": "pending",
    "error": null,
    "result": null,
    "performance": null
  }
}

Error Handling#

  • 400: Invalid input format
  • 413: File too large
  • 415: Unsupported file type
  • 429: Rate limit exceeded
  • 500: Processing error

Use Cases#

Music Production#

  • Isolate vocals for remixing
  • Extract drum patterns for sampling
  • Separate bass lines for analysis
  • Create karaoke versions

Audio Analysis#

  • Study individual instrument tracks
  • Analyze mixing techniques
  • Identify audio artifacts
  • Quality assessment

Content Creation#

  • Create backing tracks
  • Generate instrumental versions
  • Extract samples for new compositions
  • Audio restoration projects

Tips for Best Results#

Input Quality#

  • Use lossless formats when possible (WAV, FLAC)
  • Ensure proper gain staging (avoid clipping)
  • Clean, well-recorded source material
  • Consistent audio levels

Processing#

  • Allow adequate processing time
  • Check output quality before proceeding
  • Use appropriate sample rates
  • Consider file size limitations

Next Steps#

Once you have your separated stems:

  1. Review Quality - Check each stem for artifacts
  2. Process Further - Apply effects or processing
  3. Mix and Master - Recombine or use individually
  4. Export - Save in your preferred format

Ready to separate your audio? Visit the Split Audio Stems tool to get started!