How to explore this new AI tool that generates music from text prompts
Stable Audio can synthesise CD-quality audio from text in seconds.
If you are not a subscriber of The Fanbase Builder, join the dozens of artists, creators and music industry executives who receive it for free.
Let’s dive into today’s topic:
How to explore this new AI tool that generates music from text prompts
Today we add Stable Audio to the ever-growing list of music AI tools. From the creators of Stable Diffusion, Stable Audio offers an easy-to-use interface to generate music from text in seconds.
Why it matters
Many tools use generative AI techniques to create music. Since the last edition where I covered this subject, things have developed rapidly. Every new model makes music AI tools faster in delivering music of higher quality. Now, Stable Audio can synthesise CD-quality audio from text in seconds.
Big tech companies such as Meta, TikTok, and Google launched their own music AI tools. Stability AI, however, is an independent and partially open company with a proven track record of AI tools and models such as Stable Diffusion. It’s valuable to explore their approach to music AI and to spend some time with Stable Audio.
How it works
First, sign up for a free account. Stable Audio’s free version lets users create up to 45 seconds of audio for 20 tracks a month.
Writing prompts for Stable Audio is easy to learn. It is suggested to compose the prompt with keywords that describe the music’s genre, descriptive phrases, moods, used instruments, and even BPM.
For example, the prompt nineties eurodance, trance, synthesizer, 909, Juno 106, 132 bpm, warehouse rave, euphoric
generates this pumping tune:
It’s also possible to compose the prompt based on an open briefing. For example, the prompt A welcoming theme song for a newsletter about music strategy
sounds like a nice opening tune for a podcast or YouTube channel:
Or make animal sounds, like this elephant meowing
:
Go deeper by exploring Stable Audio’s user guide on writing prompts.
Yes, but..
Stable Audio offers a professional and enterprise tier that allows for commercial use. Stable Audio is trained with a dataset from the stock music licensing company AudioSparx. I’m not sure to what extent the original creators are compensated.
Take action now
Stable Audio is the best music AI tool I’ve worked with so far. Spend some time with the tool, create some music and other forms of audio with it, and come up with ideas on how artists can use these tools to their advantage,
Further reading
Announcing Stable Audio, a product for music & sound generation (Stability AI)
Stable Audio: Fast Timing-Conditioned Latent Audio Diffusion (Stability AI)
Stable Diffusion maker launches Stable Audio text-to-music AI (MusicAlly)
Stability AI releases AI audio platform (The Verge)
This AI tool generates music from text prompts (The Fanbase Builder)
This AI tool generates animations from text, images or videos (The Fanbase Builder)