Tasks API

A Task defines a set of AI processing operations to run together in a single request.

Each Task contains an array of separation targets which define which models to run the input media through. You to create a single Task that will process the same input media to produce multiple source separations or other audio processes such as transcription or alignment.

Base URL

https://api.audioshake.ai

Authentication

All of the new Tasks endpoints are authenticated via API key as provided in the AudioShake Dashboard.

Generate your key in the AudioShake Dashboard → Settings.
Add it to every request using the x-api-key header.
API keys are securely generated and can be revoked at any time.

GET /tasks
Host: api.audioshake.ai
x-api-key: <AUDIOSHAKE_API_KEY>

Create a task

Method/Path: POST /tasks
Auth: x-api-key: <AUDIOSHAKE_API_KEY>

Purpose: Create a task to process a media URL or Asset with up to 10 target operations.

Request

Headers

x-api-key: <AUDIOSHAKE_API_KEY>
Content-Type: application/json

Body shape

Field	Type	Required	Constraints	Description
`url`	`string`	no	format: uri	Publicly reachable HTTPS URL to the source audio/video. Provide either `url` or `assetId`.
`assetId`	`string`	no		ID of a previously uploaded asset. Provide either `assetId` or `url`.
`targets`	`target[]`	yes	minItems: 1; maxItems: 10	List of operations to run. 1–10 items.

Input: Provide either url or assetId. If you provide one, only that field is returned in the Task response.

Target object

Field	Type	Required	Constraints	Description
`model`	`string`	yes	Supported Models	Target model identifier.
`formats`	`string[]`	yes	Supported Formats	Desired output formats.
`residual`	`boolean`	no	default: false	Include residual audio if supported by the model. The residual is everything but the target after separation.
`variant`	`string`	no	Model Variants	Optional variants can be selected for multi_voice, vocals, and instrumental.
`language`	`string`	no	Supported Languages	Set transcription or alignment language. Adding the language attribute to your API call can help improve performance. If no language is set, it will automatically be detected.

Example Request Body:

{
  "url": "https://demos.audioshake.ai/demo-assets/shakeitup.mp3",
  "targets": [
    {
      "model": "music_removal",
      "formats": [
        "mp4"
      ]
    },
    {
      "model": "multi_voice",
      "formats": [
        "wav"
      ],
      "variant": "two_speaker"
    },
    {
      "model": "vocals",
      "formats": [
        "wav"
      ],
      "variant": "high_quality"
    }
  ]
}

cURL

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",
    "targets": [
      { "model": "dialogue", "formats": ["wav"] },
      { "model": "effects", "formats": ["wav"] },
      { "model": "multi_voice", "formats": ["wav"], "variant": "two_speaker" }
    ]
  }'

Responses

200 OK

{
  "id": "<task-id>",
  "createdAt": "2025-09-24T17:41:59.851Z",
  "clientId": "<your-client-id>",
  "url": "https://demos.audioshake.ai/demo-assets/shakeitup.mp3",
  "targets": [
    {
      "id": "<target-id>",
      "model": "dialogue",
      "status": "processing",
      "formats": ["mp4"],
      "output": [],
      "cost": 12
    },
    {
      "id": "<target-id>",
      "model": "effects",
      "status": "processing",
      "formats": ["wav"],
      "output": [],
      "cost": 6
    },
    {
      "id": "<target-id>",
      "model": "multi_voice",
      "status": "processing",
      "formats": ["wav"],
      "output": [],
      "cost": 60,
      "variant": "two_speaker"
    }
  ]
}

Response notes

Optional fields are omitted when not applicable (for example: error, variant, language, residual, and the top-level input field you did not provide).

List tasks

Method/Path: GET /tasks
Auth: x-api-key: <AUDIOSHAKE_API_KEY>

Purpose: Return a list of tasks.

Request

Headers

x-api-key: <AUDIOSHAKE_API_KEY>
Content-Type: application/json

Name	In	Type	Required	Description
clientId	path	string	yes	Client identifier.
skip	path	int	no	Start paging param
take	path	int	no	Limit paging param

cURL

curl -sS -X GET "https://api.audioshake.ai/tasks?skip=1&take=1" \
  -H "Accept: application/json" \
  -H "x-api-key: $AUDIOSHAKE_API_KEY"

Responses

200 OK

[
  {
    "id": "<task-id>",
    "createdAt": "2020-01-01T00:00:00.000Z",
    "clientId": "<client-id>",
    "targets": [
      {
        "id": "<target-id>",
        "model": "music_removal",
        "status": "processing",
        "formats": ["mp4"],
        "output": [],
        "cost": 12,
        "duration": 24
      }
    ]
  }
  // ... additional tasks
]

Get a task by id

Method/Path: GET /tasks/{id}
Auth: x-api-key: <AUDIOSHAKE_API_KEY>

Request

Headers

x-api-key: <AUDIOSHAKE_API_KEY>
Content-Type: application/json

Name	In	Type	Required	Description
id	path	string	yes	Unique identifier of the task

cURL

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

Responses

200 OK

{
  "id": "cmfy9u6m0004rowxc8zeei57j",
  "createdAt": "2025-09-24T17:41:59.851Z",
  "updatedAt": "2025-09-24T17:41:59.851Z",
  "clientId": "4448d4a8-1051-7054-a06b-5904d1c20b62",
  "targets": [
    {
      "id": "cmfy9u6m0004sowxcfqvotxxt",
      "createdAt": "2025-09-24T17:41:59.851Z",
      "updatedAt": "2025-09-24T17:49:38.781Z",
      "url": "https://example.com/demos-source/surfing.mp4",
      "model": "music_removal",
      "taskId": "cmfy9u6m0004rowxc8zeei57j",
      "status": "completed",
      "formats": ["mp4"],
      "output": [
        {
          "name": "music_removal",
          "format": "mp4",
          "type": "video/mp4",
          "link": "https://cdn.audioshake.ai/.../music_removal.mp4"
        }
      ],
      "cost": 12,
      "error": null,
      "duration": 183.12498474121094,
      "variant": "",
      "residual": false,
      "language": "en"
    }
    // ... additional targets
  ]
}

404 Resource not found

{
  "statusCode": 404,
  "error": "Not Found",
  "message": "Cannot find task"
}

Task statistics

Method/Path: GET /tasks/statistics
Auth: x-api-key: <AUDIOSHAKE_API_KEY>
Name: usage or distribution

Purpose: Aggregated task statistics (field set to be finalized).

Request

Headers

x-api-key: <AUDIOSHAKE_API_KEY>
Content-Type: application/json

Name	In	Type	Required	Description
name	path	string	yes	usage or distribution

cURL

Usage Example

curl -sS -X GET "https://api.audioshake.ai/tasks/statistics?name=usage" \
  -H "Accept: application/json" \
  -H "x-api-key: $AUDIOSHAKE_API_KEY"

Distribution Example

curl -sS -X GET "https://api.audioshake.ai/tasks/statistics?name=distribution" \
  -H "Accept: application/json" \
  -H "x-api-key: $AUDIOSHAKE_API_KEY"

Responses

200 OK Usage

[
  {
    "totalTargets": 32,
    "month": "2025-09",
    "totalMinutes": "118.04"
  }
]

200 OK Distribution

[
  { "count": 6, "model": "vocals", "percentage": "18.75" },
  { "count": 4, "model": "instrumental", "percentage": "12.5" },
  { "count": 3, "model": "piano", "percentage": "9.38" },
  { "count": 3, "model": "wind", "percentage": "9.38" },
  { "count": 3, "model": "multi_voice", "percentage": "9.38" },
  { "count": 2, "model": "drums", "percentage": "6.25" },
  { "count": 2, "model": "music_removal", "percentage": "6.25" }
  // ... additional models
]

Error Responses:

statusCode	error	message
400	Bad Request	Insufficient credits
404	Not Found	"Cannot find <task>"

Insufficient credits: The AudioShake API is restricted to accounts with sufficient credits. If you run out, you'll see an error message like this; simply return to the dashboard to purchase more.

Base URL​

Authentication​

Create a task​

Request​

Target object​

Responses​

List tasks​

Request​

Responses​

Get a task by id​

Request​

Responses​

Task statistics​

Request​

Responses​

Error Responses:​

Base URL

Authentication

Create a task

Request

Target object

Responses

List tasks

Request

Responses

Get a task by id

Request

Responses

Task statistics

Request

Responses

Error Responses: