We can’t predict what kind of music people will want to make or hear in the future. But judging from the sounds coming out of today's studios and clubs, it's a good bet that the tunes of tomorrow will be heavily mediated by digital technology.
This week’s show asks how software has changed the way composers and performers make music, and how our tools for creating music will evolve in the near future.
The germ of the episode was my interview way back in October 2016 with computer scientist Douglas Eck, the leader of a machine learning project at Google called Magenta. Eck’s team is combining techniques called deep learning and reinforcement learning to create “generative models” that can sift through enormous amounts of data and apply rules to create interesting new art and music. Magenta’s real-time generative models are already accompanying human performers in simple call-and-response settings, and Eck hops to advance them to the point that musicians will want to jam with them—and even “break” them—to see what new sounds and styles they’ll make possible.
The more I learned about Magenta, the clearer it became that the music world is on the cusp of yet another quantum leap, similar to those spurred by inventions like the piano, the phonograph, or the electric guitar. New instruments and technologies have always inspired music creators to try new things. The current wave of AI-based approaches will be no different. And while the prospect of music written by machines may be unfamiliar, Magenta hardly seemed fearsome. In fact, it seemed likely that it would push computer-savvy jazz, rock, and classical musicians in directions they could never have anticipated.
After that Google visit I gradually gathered more thread for a future-of-music episode, including interviews with classical composer Rudi Seitz, DJ and EDM label manager Biyeun Buczyk, and commercial composers Joel Roston and Andrew Willis. Each of them gave me a different way to look at the role of software in music creation. But it wasn’t until I heard a special six-part podcast series called Ways of Hearing, hosted by indie musician Damon Krukowski and co-produced by my friend Ian Coss, that I was able to identify a unifying theme and concept for my episode.
I was riveted by Ways of Hearing as well as Krukowski’s companion book, The New Analog: Listening and Reconnecting in a Digital World. Krukowsi isn’t simply offering another argument for why vinyl sounds better (though he does that too). He’s cataloguing aspects of analog sound production and transmission, such as the warm tone of an old-fashioned land-line phone call, that we can only fully appreciate now that they’ve been swept away by digital technologies.
What seemed off to me about the series and the book, though, was Krukowski's attitude of mournful nostalgia. I got the sense that he would be happier making albums with his old four-track tape deck than with Pro Tools. As a technology journalist who’s seen over and over again how innovation drives improvements in people’s lives, I don't have much patience for that kind of wistfulness.
My own feeling is that in music’s long transition from analog to digital, we’ve gained far more than we’ve lost, at least when it comes to creating and sharing music. (Whether musicianship is still a sustainable career in the digital age is a different question. I wanted to tackle it in this episode, but the drafts just got too long. I hope to come back to that question in a future episode.)
So that’s what this episode is about: the good ideas (and also a few bad ones) that people get about music creation when they’re presented with powerful new tools like synthesizers, drum machines, sequencers, digital audio workstations, and generative models. As you'll hear in the show's four segments or "movements," what we may have given up in the fidelity or richness of analog sound, we’ve gotten back a hundredfold in the flexibility, access, and creativity these digital tools afford.
MUSIC NOTE: This is the very first episode of Soonish to feature a custom, original score. It's a delightful one, and it's more than a little bit meta, as you'll hear if you listen to the episode carefully. (We recommend listening with headphones rather than earbuds or, God forbid, through your phone speaker.) It was composed by Joel Roston with editing and production help from his business partner Andrew Willis. They're featured in the episode, and they're the principals of the Boston-based music house Titlecard Music and Sound.
To help support Soonish financially, please listen to the show on RadioPublic. It's free and easy to use, and now RadioPublic pays podcasters every time you listen in the app!
Mentioned In This Episode
Canons by Rudi Seitz
DJ Beyun aka Biyeun Buczyk
The Soonish opening theme is by Graham Gordon Ramsay
Original score produced by Joel Roston and Andrew Willis of Titlecard Music and Sound
“Moonlight Serenade” by the Glenn Miller Orchestra
“Hound Dog” by Elvis Presley
“Bridge Over Troubled Water” by Simon & Garfunkel
“Hotel California” by The Eagles
“Funkytown” by Lipps Inc.
“Smells Like Teen Spirit” by Nirvana
Symphony No. 9, Movement 2 by Ludwig van Beethoven
“American Beauty” by Thomas Newman
“The Gentle Hum of Anxiety” by Trent Reznor and Atticus Ross
“Sunstone” by Rudi Seitz
“Sapphire” by Rudi Seitz
“Manic Acid (6 Days A Week)” by Beyun
“Rainbowgram” by Andrew Huang
Elsewhere at Hub & Spoke
Crimson Bikes—Culture Hustlers Episode 13
Chamber of Facts—Hi-Phi Nation Season 2, Episode 10
Special Episode: Reaching Mars—Iconography
Hiroshi Sugimoto’s Byrd Theater, Richmond, 1993 (1993)—The Lonely Palette
Shifting Blame—Ministry of Ideas, Episode 17
Biyeun Buczyk, Kip Clark, David Day, Douglas Eck, Joseph Fridman, Jason Freidenfelds, Mmmmaven, Mark Pelofsky, Joel Roston, Rudi Seitz, Daniel Sheehan, Andrew Willis