Available now

Stable Audio 2.5: text to audio

Stable Audio 2.5 makes music or sound effects from a written description, from one second up to 190 seconds long.

Model ID

Input

Fields marked * are required. Review the price before generating.

Example prompt. Edit it or write your own.

Additional settings

The duration of the audio clip to generate

The number of steps to denoise the audio for

How strictly the diffusion process adheres to the prompt text (higher values make your audio closer to your prompt).

Price$0.32 / run

You sign up first. Your settings are kept.

About this model

Overview

Stable Audio 2.5 turns a text description into audio: a music cue, an ambient bed or a single sound effect. Set the length in whole seconds, up to 190.

Guidance sets how closely the result follows your words, and a seed lets you repeat a result you liked. Fewer steps give a rougher draft; the default uses the most.

What you can build

  • Background music for videos and podcasts
  • Ambient loops for games and installations
  • Short sound effects for edits and apps
  • Rough music sketches to test a mood

Prompting tips

  • Name genre, instruments, tempo in BPM and mood, for example "lo-fi hip hop, warm piano, 80 BPM, calm".
  • Set the length to what you need; a short effect does not need the full 190 seconds.
  • Keep the seed when you like a result and change one word of the description at a time.

What to check

Listen for abrupt endings, unwanted sounds and changes that do not fit the brief.

Inputs

  • Prompt (required): The prompt to generate audio from.
  • Seconds Total (optional): The duration of the audio clip to generate. From 1 to 190.
  • Number of Inference Steps (optional): The number of steps to denoise the audio for. From 4 to 8.
  • Guidance Scale (optional): How strictly the diffusion process adheres to the prompt text (higher values make your audio closer to your prompt). From 1 to 25.
  • Seed (optional)

Output and limits

  • Sung or spoken words can come out unclear or differ from your text.
  • Exact length, tempo and key are not guaranteed; listen to the whole result before you use it.
  • Prompts that name real artists or protected songs can be refused.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • up to, default options: $0.32 / run