AI chatbots and image generators are all the craze, and soon, AI audio generators may join that list.
On Wednesday, Stability AI released Stable Audio, which uses generative AI techniques to create music, sound effects, and audio up to 90 seconds long based on user prompts. Stable Audio’s release follows similar efforts from other industry giants like Google and Meta.
Also: Adobe Firefly, now out of beta, boasts fix for DALL-E’s drawbacks
Users can test out the technology themselves by using the basic free version, which enables the generation of up to 20 45-second non-commercial tracks a month.
If you plan on using this tool as part of your workflow, consider investing in the Pro subscription, which delivers up to 500 90-second long tracks per month with a commercial license.
Stability AI said that the AI tool will be especially useful for musicians who can use it to create samples to incorporate into their music.
“Our hope is that Stable Audio will empower music enthusiasts and creative professionals to generate new content with the help of AI, and we look forward to the endless innovations it will inspire,” said Emad Mostaque, CEO of Stability AI.
I tried the tool for myself, and here’s how it went.
How to access Stable Audio
To try it out for yourself, all you need to do is visit the Stable Audio website and create or sign into an account. Because the tool is free, a credit card is not required.
Currently, because of the high interest in — and traffic volume to — the application, you may be greeted by an error message or blank page when going to the site.
Also: How Apple weaved AI into the iPhone 15 and Apple Watch 9
Stability AI acknowledged the issue on X (formerly Twitter), noting that the high demand pushed their servers to total capacity. If you can’t get in, they recommend you try again in 24 hours.
When I was met with the error message, I just kept trying to access the site on different incognito browsers, and eventually, it went through. I don’t know if that’s a foolproof approach but it’s worth a try since it worked for me.
Generating Music
Once you are in, all you need to do is type in a prompt for the sound you’d like to be generated, and after about 15 seconds, you can expect your result.
For writing the prompt, you can include as much detail as you’d like to get the best results, including details such as genre, mood, length, BPM, and more.
Also: The 10 best ChatGPT plugins of 2023 (and how to make the most of them)
Stability AI’s example prompts included “Post-Rock, Guitars, Drum Kit, Bass, Strings, Euphoric, Up-Lifting, Moody, Flowing, Raw, Epic, Sentimental, 125 BPM” or something as simple as “car passing by.”
I used the prompt “Happy, beachy, mellow country track,” as seen by the photo above, and was impressed with the output, which had the exact vibe I was searching for. Once the track is generated, you have the option to download the track as an mp3 file.