In March, we saw the launch of a “ChatGPT for music” called Suno, which uses generative AI to provide realistic songs on demand from short text prompts. Just a few weeks later, an analogous competitor—Udio—arrived on the scene.
I’ve been working with various creative computational tools for the past 15 years, each as a researcher and a producer, and the recent pace of change has floored me. As I’ve argued elsewhere, the view that AI systems won’t ever make “real” music like humans do needs to be understood more as a claim about social context than technical capability.
The argument “sure, it could actually make expressive, complex-structured, natural-sounding, virtuosic, original music which might stir human emotions, but AI can’t make proper music” can easily begin to sound like something from a Monty Python sketch.
After fidgeting with Suno and Udio, I’ve been eager about what it is strictly they modify—and what they may mean not just for the best way professionals and amateur artists create music, but the best way all of us eat it.
Expressing Emotion Without Feeling It
Generating audio from text prompts in itself is nothing latest. Nevertheless, Suno and Udio have made an obvious development: from an easy text prompt, they generate song lyrics (using a ChatGPT-like text generator), feed them right into a generative voice model, and integrate the “vocals” with generated music to provide a coherent song segment.
This integration is a small but remarkable feat. The systems are superb at making up coherent songs that sound expressively “sung” (there I’m going anthropomorphizing).
The effect might be uncanny. I realize it’s AI, however the voice can still cut through with emotional impact. When the music performs a superbly executed end-of-bar pirouette right into a latest section, my brain gets a few of those little sparks of pattern-processing joy that I would get listening to a terrific band.
To me this highlights something sometimes missed about musical expression: AI doesn’t must experience emotions and life events to successfully express them in music that resonates with people.
Music as an On a regular basis Language
Like other generative AI products, Suno and Udio were trained on vast amounts of existing work by real humans—and there’s much debate about those humans’ mental property rights.
Nevertheless, these tools may mark the dawn of mainstream AI music culture. They provide latest types of musical engagement that individuals will just wish to use, to explore, to play with, and truly hearken to for their very own enjoyment.
AI able to “end-to-end” music creation is arguably not technology for makers of music, but for consumers of music. For now it stays unclear whether users of Udio and Suno are creators or consumers—or whether the excellence is even useful.
A protracted-observed phenomenon in creative technologies is that as something becomes easier and cheaper to provide, it’s used for more casual expression. Because of this, the medium goes from an exclusive high art form to more of an on a regular basis language—think what smartphones have done to photography.
So imagine you can send your father a professionally produced song all about him for his birthday, with minimal cost and energy, in a form of his preference—a modern-day birthday card. Researchers have long considered this eventuality, and now we will do it. Glad birthday, Dad!
Can You Create Without Control?
Whatever these systems have achieved and should achieve within the near future, they face a glaring limitation: the shortage of control.
Text prompts are sometimes not much good as precise instructions, especially in music. So these tools are fit for blind search—a sort of wandering through the space of possibilities—but not for accurate control. (That’s to not diminish their value. Blind search could be a powerful creative force.)
Viewing these tools as a practicing music producer, things look very different. Although Udio’s about page says “anyone with a tune, some lyrics, or a funny idea can now express themselves in music,” I don’t feel I actually have enough control to precise myself with these tools.
I can see them being useful to seed raw materials for manipulation, very similar to samples and field recordings. But after I’m in search of to precise myself, I want control.
Using Suno, I had some fun finding essentially the most gnarly dark techno grooves I could get out of it. The result was something I might absolutely use in a track.
But I discovered I could also just gladly listen. I felt no compulsion so as to add anything or manipulate the result so as to add my mark.
And plenty of jurisdictions have declared that you just won’t be awarded copyright for something simply because you prompted it into existence with AI.
For a start, the output depends just as much on every thing that went into the AI—including the creative work of thousands and thousands of other artists. Arguably, you didn’t do the work of creation. You just requested it.
Latest Musical Experiences within the No-Man’s Land Between Production and Consumption
So Udio’s declaration that anyone can express themselves in music is an interesting provocation. The individuals who use tools like Suno and Udio could also be considered more consumers of music AI experiences than creators of music AI works, or as with many technological impacts, we may have to give you latest concepts for what they’re doing.
A shift to generative music may draw attention away from current types of musical culture, just because the era of recorded music saw the diminishing (but not death) of orchestral music, which was once the one technique to hear complex, timbrally wealthy and loud music. If engagement in these latest varieties of music culture and exchange explodes, we might even see reduced engagement in the normal music consumption of artists, bands, radio and playlists.
While it is just too early to inform what the impact might be, we needs to be attentive. The trouble to defend existing creators’ mental property protections, a big moral rights issue, is a component of this equation.
But even when it succeeds I imagine it won’t fundamentally address this potentially explosive shift in culture, and claims that such music may be inferior even have had little effect in halting cultural change historically, as with techno and even jazz, way back. Government AI policies may have to look beyond these issues to grasp how music works socially and to be sure that our musical cultures are vibrant, sustainable, enriching, and meaningful for each individuals and communities.
