The person responsible for artificial intelligence creating music is the company Stability AI and we cannot say that it sounds bad.
Artificial intelligence, or simply called AI by its initials, is on everyone’s lips and every day more. Many things have already been said about this revolutionary technology, including that it will jeopardize the jobs of millions of people, that it has a large environmental footprint and that it has a dark side. Apart from this, AI can also compose music and here we have to say that things don’t sound bad.
In a recent article published by Ars Technica has echoed the work that the company Stability AI is doing with the Stable Audio AI model. This artificial intelligence program is able to synthesize music or sounds from written descriptions. A technology that, according to Ars Technica, could challenge human musicians.
A new AI allows you to generate music with good quality
Stability AI is not an unknown company, as it helped finance the creation of Stable Diffusion. Another artificial intelligence model that was used to synthesize images. This group and another called Harmonai collaborated to develop a music generator. Now, after a lot of work, Stable Audio is released on the market.
In their website, the companies have left samples of this musical production made by AI. The truth is that they don’t sound bad at all and They are a significant improvement compared to the quality that users were accustomed to seeing on the Internet.
Stability has communicated that it has the idea that Stable Audio Available in a free tier and a $12 monthly Pro plan. With the free option, those interested will be able to generate 20 tracks per month with a maximum duration of 20 seconds. As for the Pro plan, 500 generations will be allowed with durations of up to 90 seconds.
As the article points out future versions are expected to include open source models based on Stable Audio, as well as training codes for those who want to develop audio generation models. Stability will have to be given time to see how these two issues evolve.
Although there has been controversy over the use of AI in graphic arts and entertainment, the truth is that it is still a quite suggestive (and cheap) resource to apply on different levels. For example, for podcasts and YouTube videosadd royalty-free intro, ending or background music.
Apart from this, it is true that improper use of this technology could, and is already causing, problems for artists and beyond. And AI is changing things very quickly. A recent case that went viral on Twitter Spain was a music channel that uploaded famous songs covered with the voice of Peter Griffin’s animated character from the Family Guy series.
At the moment AIs remain easily identifiable models, but they are becoming more and more refined in the work they do. We will have to see what happens in the not so distant future., whether governments will implement limitations on its use or openly pursue it under certain conditions. Time will tell.