Stable Audio is a platform that allows users to create original music and sound effects using artificial intelligence. The platform uses audio diffusion AI models to generate music and sound effects based on text prompts provided by the user.
Table of Contents
Quick guide
Here’s a guide to help you get started with the free version:
Click on ‘Try it out’
> Left bottom of your screen:
2. Create an account
3. Log-in
Generate audio:
Enter a prompt and set the duration of your audio-track:
Prompts are the text you use to tell the AI how you want your music to sound. You can include details such as: genres, descriptive phrases, instruments, moods, and beats per minute (BPM).
Example
"Cinematic, Soundtrack, Wild West, High Noon Shoot Out, Percussion, Whistles, Horses, Action Scene, SFX, Shaker, Guitar, Bass, Timpani, Strings, Tense, Climactic, Atmospheric, Moody".
Optional: Select a ‘Prompt Library’ for inspiration on different genres:
3. Click on ‘Generate’
Extra options
Advanced users can adjust three (extra) settings in the free version of Stable Audio:
Steps: The ‘steps’ refer to how many times the AI goes over the data and adds detail to it. The number of steps can affect the quality of the output. Too few steps might result in an output that looks thrown together, while too many steps can lead to diminishing returns and risk adding too much detail.
Seed: The ‘seed’ is an input that informs the specific arrangement of noise used to generate your audio. By default, this input is set to ‘random’. However, if you provide your own input (e.g., 222222), it will result in a specific audio arrangement.
Prompt Strength: The ‘prompt strength’ option controls how closely the model attempts to guide the audio to your text prompt. This means that a higher prompt strength will result in the generated audio more closely matching the details specified in your text prompt. Conversely, a lower prompt strength allows for more variation and less adherence to the prompt. This can be useful if you want to experiment with different sounds or if you’re looking for a more unexpected result.
Source: Stable Audio — User Guide
Edit audio:
Stable Audio also lets you edit the soundtracks that you have created:
Select the soundtrack in the overview of previously created audio:
2. Click on ‘Use as input’:
3. Enter a ‘Prompt’ that describe the changes you want to make to the audio
4. Adjust the ‘Input strength’ of your prompt:
5. Click on ‘Generate’