AI music production is controversial. There are many musicians who feel that the onset of AI won’t end human art, but spur a new golden era of creativity. But there are others who believe that using AI will lead to the death of music. I’m here to tell you, as a sound designer, that such fears are unfounded.
I have worked in sound for over twenty years now. I’ve done sound design for film and video games, performed as a drummer and recording engineer, and even taught a few online courses on the physics of music production. I also have an M.D in computer science from UoPX where I wrote my thesis on applying machine learning techniques to music production. Nowadays, AI is being used in all sorts of creative pursuits; it is often used to generate images and text by using algorithms to learn patterns from huge data sets of existing art.
Music is no exception; machine learning algorithms have been applied to music for decades now with varying degrees of success. But more recently, computer scientists are making great strides in “reverse engineering” existing songs into code – converted into structured data which can be analyzed by computers – giving us the ability to generate new scores that show striking similarities with existing songs without relying on human input at all. Some artists are even using these tools creatively; instead of just generating new content based on existing material they can use this technology to explore new musical styles or even collaborate with AI-generated artists whose musical DNA they find appealing (while still retaining creative control).
As an artist myself, this technology is quite exciting – I’m always looking for an edge over my competition, and while it might seem like an abuse of my skills as an audio engineer or composer it could actually make me better at what I do by giving me new tools and techniques for enhancing my own work or extending the boundaries of what is possible through automation rather than brute force human effort alone.
There are many different ways this technology can be used in audio production, but for the rest of thisarticle, I'm going to talk about why we should use it.
The main reason why AI is so powerful in music production, in my mind, is that it gives musicians access to melodies, sounds, and timing signatures they would have never been able to imagine otherwise. If you think about it, if you were trying to create a song with human musicians, you would need other humans to play every single part. We ideate and innovate based on the sounds of others - that's part of what music is all about.
But with artificial intelligence, you can offload this task to computers. You don’t need a drummer, pianist, DJ, or producer; you simply need an algorithm capable of producing the right waveform. With future music production algorithms, you'll be able to give a vague description of how you want the drums set up and it'll spit you back an accompanying track.
This gives artists access to a whole new world of sound and creativity – sounds they might not have thought about otherwise - and it serves as a springboard for their other ideas!
It also means that artists can focus on other aspects of making great music – like writing lyrics or rocking out on the guitar – without having to worry as much about getting every little detail down perfectly.