Nearly 200 years after his death, the German composer’s musical scratch was pieced together by machine—with a lot of human help.
Teresa Carey: This is Scientific American’s 60-Second Science. I'm Teresa Carey.
Every morning at five o’clock, composer Walter Werzowa would sit down at his computer to anticipate a particular daily e-mail. It came from six time zones away, where a team had been working all night (or day, rather) to draft Beethoven’s unfinished 10th Symphony—almost two centuries after his death.
The e-mail contained hundreds of variations, and Werzowa listened to them all.
Werzowa: So by nine, 10 o’clock in the morning, it’s like I’m already in heaven.
Carey: Werzowa was listening for the perfect tune—a sound that was unmistakably Beethoven.
But the phrases he was listening to weren’t composed by Beethoven. They were created by artificial intelligence—a computer simulation of Beethoven’s creative process.
Werzowa: There were hundreds of options, and some are better than others. But then there is that one which grabs you, and that was just a beautiful process.
Carey: Ludwig van Beethoven was one of the most renowned composers in Western music history. When he died in 1827, he left behind musical sketches and notes that hinted at a masterpiece. There was barely enough to make out a phrase, let alone a whole symphony. But that didn’t stop people from trying.
In 1988 musicologist Barry Cooper attempted. But he didn’t get beyond the first movement. Beethoven’s handwritten notes on the second and third movements are meager—not enough to compose a symphony.
Werzowa: A movement of a symphony can have up to 40,000 notes. And some of his themes were three bars, like 20 notes. It’s very little information.
Carey: Werzowa and a group of music experts and computer scientists teamed up to use machine learning to create the symphony. Ahmed Elgammal, the director of the Art and Artificial Intelligence Laboratory at Rutgers University, led the AI side of the team.
Elgammal: When you listen to music read by AI to continue a theme of music, usually it’s a very short few seconds, and then they start diverging and becoming boring and not interesting. They cannot really take that and compose a full movement of a symphony.
Carey: The team’s first task was to teach the AI to think like Beethoven. To do that, they gave it Beethoven’s complete works, his sketches and notes. They taught it Beethoven's process—like how he went from those iconic four notes to his entire Fifth Symphony.
[CLIP: Notes from Symphony no. 5]
Carey: Then they taught it to harmonize with a melody, compose a bridge between two sections and assign instrumentation. With all that knowledge, the AI came as close to thinking like Beethoven as possible. But it still wasn’t enough.
Elgammal: The way music generation using AI works is very similar to the way, when you write an e-mail, you find that the e-mail thread predicts what’s the next word for you or what the rest of the sentence is for you.
Carey: But let the computer predict your words long enough, and eventually, the text will sound like gibberish.
Elgammal: It doesn’t really generate something that can continue for a long time and be consistent. So that was the main challenge in dealing with this project: How can you take a motif or a short phrase of music that Beethoven wrote in his sketch and continue it into a segment of music?
Carey: That’s where Werzowa’s daily e-mails came in. On those early mornings, he was selecting what he thought was Beethoven’s best. And, piece by piece, the team built a symphony.
Matthew Guzdial researches creativity and machine learning at the University of Alberta. He didn’t work on the Beethoven project, but he says that AI is overhyped.
Guzdial: Modern AI, modern machine learning, is all about just taking small local patterns and replicating them. And it’s up to a human to then take what the AI outputs and find the genius. The genius wasn’t there. The genius wasn’t in the AI. The genius was in the human who was doing the selection.
Carey: Elgammal wants to make the AI tool available to help other artists overcome writer’s block or boost their performance. But both Elgammal and Werzowa say that the AI shouldn’t replace the role of an artist. Instead it should enhance their work and process.
Werzowa: Like every tool, you can use a knife to kill somebody or to save somebody’s life, like with a scalpel in a surgery. So it can go any way. If you look at the kids, like kids are born creative. It’s like everything is about being creative, creative and having fun. And somehow we’re losing this. I think if we could sit back on a Saturday afternoon in our kitchen, and because maybe we’re a little bit scared to make mistakes, ask the AI to help us to write us a sonata, song or whatever in teamwork, life will be so much more beautiful.
Carey: The team released the 10th Symphony over the weekend. When asked who gets credit for writing it— Beethoven, the AI or the team behind it—Werzowa insists it is a collaborative effort. But, suspending disbelief for a moment, it isn’t hard to imagine that we’re listening to Beethoven once again.
Werzowa: I dare to say that nobody knows Beethoven as well as the AI, did—as well as the algorithm. I think music, when you hear it, when you feel it, when you close your eyes, it does something to your body. Close your eyes, sit back and be open for it, and I would love to hear what you felt after.
Carey: Thanks for listening. For Scientific American’s 60-Second Science, I’m Teresa Carey.
[The above text is a transcript of this podcast.]