Skip to main content

Getting Started with Tasks API NEW

The AudioShake Tasks API lets you submit a single job that runs multiple audio operations—such as stem separation, music removal, or lyric transcription—in one request. This modernized API completely replaces older older, model-specific legacy endpoints with a single, unified workflow.

1. Create an Account and Get Your API Key​

  1. Visit the AudioShake Dashboard.
  2. Create an account and navigate to Settings → API Keys.
  3. Authenticate with your API key using x-api-key with your api key in each request header.
x-api-key: <YOUR_AUDIOSHAKE_API_KEY>

Basic Example: This task returns a high quality variant of the vocals and piano source stems.

Note: Vocals and Instrumental are the only high_quality modele variants.

curl -sS -X POST "https://api.audioshake.ai/tasks" \\
  -H "x-api-key: $AUDIOSHAKE_API_KEY" \\
  -H "Content-Type: application/json" \\
  -d '{
  "url": "<https://example.com/demos-source/surfing.mp4>",
  "callbackUrl": "$WEBHOOK_CALLBACK_URL",
  "targets": [
    {
      "model": "vocals",
      "formats": [
        "mp4"
      ],
        "variant": "high_quality"
    },
    {
      "model": "piano",
      "formats": [
        "mp4"
      ],
    }
  ]
}'

2. Submit Your First Task​

Each Task defines: • A source URL (audio or video) • Up to 10 target operations (e.g., vocals, music_removal, alignment)

    curl -sS -X POST "https://api.audioshake.ai/tasks" \
  -H "x-api-key: $AUDIOSHAKE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com/audio/song.mp3",
    "targets": [
      { "model": "vocals", "formats": ["wav"] },
      { "model": "instrumental", "formats": ["wav"] }
    ]
  }'

3. Check Task Progress​

Tasks are processed asynchronously. You can poll task status or register a webhook to receive notifications.

curl -sS -X GET "https://api.audioshake.ai/tasks/<TASK_ID>" \
  -H "x-api-key: $AUDIOSHAKE_API_KEY"

4. Retrieve Results​

Completed tasks return download URLs for each separated stem, transcription, or alignment file.

{
  "status": "completed",
  "targets": [
    {
      "model": "vocals",
      "output": [
        { "link": "https://cdn.audioshake.ai/.../vocals.wav" }
      ]
    }
  ]
}

5. Next Steps​

• Explore available models in Resources → Models • Use callback URLs to automate results collection • Monitor credits and usage in the AudioShake Dashboard

⸻

Need help? Contact support@audioshake.ai