- Creative AI
- Posts
- Stability AI launches new AI tool to make music; Artists oppose the tech
Stability AI launches new AI tool to make music; Artists oppose the tech
Erik Knobl
April 07, 2024
Good morning, humans.
Stability AI launches its Stable Audio 2.0 model, revolutionizing the way we create and enjoy sound. But at the same time, important musicians are raising their voices of alarm to the AI sector.
In today's Newsletter we include:
New AI tools for making music.
Stability AI improves its audio creation with Stable Audio 2.0
Stability AI, the company known for its open AI image generation models, surprises us again. This time they are diving into the world of music with the launch of Stable Audio 2.0, a model capable of creating complete and high-quality musical tracks.
Stable Audio is a free model that can generate complete instrumental songs of up to 3 minutes in length, with just a text prompt. In addition, it can receive sound samples and use them as a basis for generating the song.
How to use Stable Audio?
Go to the website stableaudio.com, and create an account:
Screenshot of Stable Audio Landing page
In the Prompt field, enter the description of the song you want to generate:
Screenshot of Stable Audio page to generate a song.
And that's it, your song will be generated!
Stable Audio has several plans. The free plan currently allows you to generate 10 songs per month.
Screenshot of the first part of the letter
Musicians raise their voices: A letter to the AI industry
More than 200 renowned artists including Billie Eilish, Katy Perry, Stevie Wonder, and Juanes signed a letter urging AI companies to be careful in their creations. They recognize the potential of AI to "boost human creativity" but oppose any technology that replaces humans or undermines their earnings.
Leading companies like Suno AI, which generate surprisingly high-quality music with musical tracks, lyrics and melodies included with just a few instructions, are in the crosshairs, as some of their creations are too similar to copyrighted material, raising suspicions that they are training their models without the corresponding licenses. For example:
Ed Newton-Rex claims to have analyzed some of Suno's results, and found that it generates music that resembles known and copyrighted melodies.
Why is it important? Beyond Suno AI's practices, it is possible that there will be a wave of legal battles related to music and AI starting in 2024, similar to what is happening with OpenAI and Midjourney in their respective fields.
While the legal landscape is being defined, users can generate songs and experiment with these powerful tools, and discover new uses.
More creative tools news
Apple announces its WWDC event for June 10-14 and says it will be (A)bsolutly (I)ncredible. We'll see if they (finally) present something related to AI.
Adobe introduced GenStudio, its generative AI platform focused on businesses, brands and marketing.
Canva buys Affinity to complement its design suite with Photoshop-like features
Do these notes help you? Are they interesting? I would appreciate any feedback you can give me to improve.
Thanks for reading. See you next week!
Hi 👋 I'm Erik Knobl, Product Designer by day and Generative AI explorer on weekends. I share my learnings in this newsletter. Consider subscribing to stay in touch.