Ace Step Prompt To Audio

Usage-based pricing
ae_ace_step_prompt_to_audio

Generate audio from a prompt using the ACE-Step model. This is an asynchronous AE tool: the call returns an AE task_id. Use ae_get_result with that same task_id to check progress and retrieve the result. If the task is still running, query the same task_id again; do not resubmit the generation request. seed is optional and accepts only an integer or null. Omit seed or set it to null to use a random seed; the string "random" is not supported.

Audio
Usage-based pricing

Pricing

The charge per call is (max(0.24 × Successfully generated audio duration (seconds), 0)) AE credits. All applicable conditions are combined to calculate the total.

Billing is based on successfully generated audio duration. Instrumental and other generation settings use the same rate. Server errors and queue wait time are not charged.

Parameters

Optional parameters use the platform defaults when omitted.

ParameterTypeRequiredDescriptionDefault
request_bodyobjectYesNo parameter description.—
request_body.seedinteger | nullNoRandom seed for reproducibility. If not provided, a random seed will be used.—
request_body.promptstringYesPrompt to control the style of the generated audio. This will be used to generate tags and lyrics.—
request_body.durationnumberNoThe duration of the generated audio in seconds.60
request_body.schedulerenum (euler | heun)NoScheduler to use for the generation process.euler
request_body.instrumentalbooleanNoWhether to generate an instrumental version of the audio.false
request_body.guidance_typeenum (cfg | apg | cfg_star)NoType of CFG to use for the generation process.apg
request_body.guidance_scalenumberNoGuidance scale for the generation.15
request_body.number_of_stepsintegerNoNumber of steps to generate the audio.27
request_body.granularity_scaleintegerNoGranularity scale for the generation process. Higher values can reduce artifacts.10
request_body.guidance_intervalnumberNoGuidance interval for the generation. 0.5 means only apply guidance in the middle steps (0.25 * infer_steps to 0.75 * infer_steps)0.5
request_body.tag_guidance_scalenumberNoTag guidance scale for the generation.5
request_body.lyric_guidance_scalenumberNoLyric guidance scale for the generation.1.5
request_body.minimum_guidance_scalenumberNoMinimum guidance scale for the generation after the decay.3
request_body.guidance_interval_decaynumberNoGuidance interval decay for the generation. Guidance scale will decay from guidance_scale to min_guidance_scale in the interval. 0.0 means no decay.0

Invocation example

Copy the request and explore its response example.

POST/agent-api/v1/tool/executeX-Api-Key

Submit first, then query with the returned task_id until the task completes.

Request coderequest.sh
curl --request POST \
  --url 'https://agentearth.ai/agent-api/v1/tool/execute?tool_name=ae_ace_step_prompt_to_audio' \
  --header "X-Api-Key: $AGENTEARTH_API_KEY" \
  --header 'Content-Type: application/json' \
  --data-raw '{
  "params": {
    "request_body": {
      "duration": 240,
      "granularity_scale": 10,
      "guidance_scale": 12,
      "guidance_type": "apg",
      "instrumental": true,
      "number_of_steps": 50,
      "prompt": "Minimal cinematic ambient underscore, dark, restrained and extremely sparse. Deep sub bass drone around 55 Hz, slowly evolving low synth pads, airy filtered noise texture, gentle tension. No drums, no percussion, no beat, no melody, no vocals, no piano. Documentary science narration background music, subtle unease, very quiet dynamics, seamless evolving, soft fade at the end.",
      "scheduler": "euler",
      "seed": 20261111
    }
  }
}'
Response exampleresponse.json
JSON
{
  "description": "task accepted; poll the result via query_tool with this task_id",
  "query_tool": "ae_get_result",
  "status": "created",
  "task_id": "efae7479-899c-414a-af29-d7b6e3f2a2d8"
}