Get API Key Get API
Docs

Add stem (add stem)

- Layer a stem onto an existing track.

  • Reference the source track: specified by task_id + audio_index, no need to record any extra id
  • Async task: submitting returns a task_id; poll GET /v1/music/tasks/:task_id for the result
Info

Source Audio Reference: No additional IDs are needed for tracks based on existing songs. Just pass task_id (the task ID from the previous run of this task) + audio_index (which result track in the data.result.music[] list, 1-based, default is 1).

Warning

custom determines which fields take effect. All submitted fields must still meet their type, range, and length requirements, even when they do not take effect in the current mode. With custom=true, prompt (lyrics), title, tags, negative_tags, style_weight, weirdness, and audio_weight take effect, while gpt_description is ignored. With custom=false, only gpt_description is used; it is required, and omitting it returns 400 during submission. If custom is omitted, the backend infers it in this order: prompt present → true; no prompt but gpt_description present → false; otherwise, tags or title present → true.

Request example

curl -X POST https://api.openveer.com/v1/music/generations/addStem \
  -H 'Authorization: Bearer <token>' \
  -H 'Content-Type: application/json' \
  --data '{
    "model": "suno",
    "task_id": "task_01ABC",
    "audio_index": 1,
    "version": "v6",
    "custom": false,
    "gpt_description": "join the warm string section"
  }'

Response example

{
  "code": 200,
  "data": [
    {
      "status": "submitted",
      "task_id": "task_01K8SGYNNNVBQTXNR4MM964S7K"
    }
  ]
}

Authorizations

Authorization string

All endpoints require authentication using a Bearer Token

Get an API Key:

Visit the API Key management page to get your API Key

Add the following to the request headers when using it:

Authorization: Bearer YOUR_API_KEY

Body

model string

Audio model. Currently pass suno (defaults to suno if omitted).

task_id string

Our task_id of the job that produced the source track. If missing or the source cannot be resolved, a 400 is returned at submission time.

audio_index integer

Selects a track from the source task's data.result.music[] array (1-based; default 1).

version string

Public version: v6 / v6-wild / v6-mini. Defaults to v6; omit this field when using custom_model_id.

custom_model_id string

The full UUID returned by the model-creation task. Mutually exclusive with version.

custom boolean

true=custom mode (prompt used as lyrics); false=inspiration mode (uses gpt_description); if omitted, inferred from the content (see the Warning above).

prompt string

Lyrics. Takes effect when custom=true (ignored in inspiration mode).

gpt_description string

Inspiration prompt. Required when custom=false — if missing, the request fails with 400 at submission (nothing is charged).

title string

Title. Only takes effect when custom=true.

tags string

Style tags. Only takes effect when custom=true.

negative_tags string

Style tags to exclude. Only takes effect when custom=true.

style_weight number

Style weight, 0.00–1.00 (out-of-range values return 400 directly at submission time). Only takes effect when custom=true.

weirdness number

Creativity weight, from 0.00 to 1.00. weirdness_constraint is accepted as a legacy alias; use weirdness for new requests.

audio_weight number

Audio weight, 0.00–1.00. Only takes effect when custom=true.

variety string

Style variation: off / normal / high / extra / max.

max_mode boolean

Enables Max mode. Requires custom mode and is billed at twice the standard price.

audio_format string

Output audio format: mp3 / m4a / wav. This endpoint does not support a target-duration field.

Note

This interface does not support the fields auto_lyrics, persona_id, instrumental, or vocal_gender, or target duration field.

Note

Get Result: This is an asynchronous task. After submission, you will receive task_id and then poll GET /v1/music/tasks/{task_id} every 3–5 seconds until status becomes completed or failed (music generation usually takes 30–120 seconds; while generation is in progress, status may be pending or processing, and progress is an integer from 0 to 100 and is not guaranteed to change at fixed intervals). Once completed, read audio_url from data.result.music[]. In case of failure, data.error.message provides the reason and the deducted amount will be automatically refunded.

Response

code integer

Response status code

data array
Returned data array
status string

Task status

  * submitted - Submitted
task_id string

Unique task identifier (used to poll GET /v1/music/tasks/{task_id} for the result)

This documentation is built and hosted on Mintlify, a developer documentation platform.