Generate music

Detailed description:
song_section
Specifies the start and end positions of the generated section within the full song. 0% (0) corresponds to the beginning of the song, 50% (0.5) corresponds to the middle of the song, and 100% (1) corresponds to the end of the song. Clicking the value indicator switches to automatic mode, letting the model determine the best placement. This is particularly useful when using the song extension feature.
lyrics_placement
Specifies the placement of lyrics within the full song. 8% corresponds to the beginning of the song, 50% corresponds to the middle of the song, and 90% will generate a section at the end of the song. This is particularly useful when using the song extension feature.
prompt_strength
Controls the influence of the prompt on the generated content. Higher values may improve adherence to the prompt, but may result in less natural sounding content.
lyrics_strength
Controls the influence of the lyrics on the generated content. Lower values may produce a more natural sound, but may ignore some of the lyrics.
clarity_strength
Controls the priority the model places on clarity and instrument separation when generating audio. Higher values may produce clearer audio, but may sound less natural.
generation_quality
Experimental feature. Used to adjust the tradeoff between generation quality and speed.

Request

Header Params

Body Params application/json

Example

{
  "gen_params": {
    "prompt": "The last song ever sung on the planet earth",
    "lyrics": "",
    "lyrics_type": "generate",
    "bypass_prompt_optimization": false,
    "seed": -1,
    "song_section_start": 0.4,
    "prompt_strength": 0.5,
    "clarity_strength": 0.25,
    "lyrics_strength": 0.5,
    "generation_quality": 0.75,
    "negative_prompt": "",
    "model_type": "udio32-v1.5",
    "config": {
      "mode": "regular"
    }
  }
}

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

Request

Request Code Samples

Responses