- A lonely grandma sought family to spend Christmas with on Craigslist Saturday 5:45 PM
- Airbnb bans white supremacists tied to Iron March forum Saturday 5:07 PM
- Did a Twitter user really get tricked into naming baby ‘Jack Ingof’? Saturday 4:46 PM
- State of emergency declared in New Orleans following ‘cyberattack’ Saturday 4:12 PM
- Video shows boy getting beat up–mom says it’s because he wore MAGA hat Saturday 3:54 PM
- Billboard changing albums chart to count YouTube streams Saturday 2:43 PM
- TikTok’s 20 most popular songs of 2019 Saturday 2:14 PM
- Greek gods memes are flooding Reddit thanks to TV reboot rumors Saturday 1:47 PM
- Anti-impeachment protesters aimlessly fumble through halls of Congress Saturday 12:54 PM
- Everything we know so far about the Xbox Series X Saturday 12:17 PM
- ASMR YouTuber Life with MaK says she was branded a ‘Nazi’ by online smear campaign Saturday 10:46 AM
- Voters duped by fake ex-Bloomberg intern’s tweet about being fired Saturday 9:47 AM
- HBO’s ‘Watchmen’ and the fantasy of competence Saturday 8:00 AM
- Cómo ver Kamaru Usman vs. Colby Covington en el UFC 245 Saturday 7:00 AM
- ‘Penis fish’ memes erupt after worms wash up on California coast Friday 5:58 PM
Artificial intelligence has reached a new frontier: creating black metal music without the need for actual musicians.
Two musical technologists named Zack Zukowski and CJ Carr have created an algorithm that can learn bits of existing music and then duplicate it to create a completely new song, the Outline writes. To prove it, Zukowski and Carr, under the name Dadabots, created a heavy metal album called Coditany of Timeness that sounds like a real metal album. That’s because it is a real metal album—just one created by AI.
Zukowski and Carr took small pieces of a 2011 album called Diotima by the death metal band Krallice, and, as the Outline explains, “Then they fed each segment through a neural network—a type of artificial intelligence modeled loosely on a biological brain—and asked it to guess what the waveform of the next individual sample of audio would be. If the guess was right, the network would strengthen the paths of the neural network that led to the correct answer, similar to the way electrical connections between neurons in our brain strengthen as we learn new skills.”
At first, Carr said, “the kinds of sounds it [produced were] very noisy and grotesque and textural.” But as the algorithm learned, the music began to sound more like Krallice. “As it improves its training, Carr said, “you start hearing elements of the original music it was trained on come through more and more.”
Here’s what it sounds like.
In reality, it’s difficult to know that the sounds were not created by dudes wearing all black and sporting scary-looking facepaint. That includes the album artwork and the names of the songs and the album, all of which were determined by the AI.
For Dadabots, it’s not all about black metal. It’s also about metalcore, as the pair used the Dillinger Escape Plan album Calculating Infinity to create an album of metal stops, starts, screams, and abrupt time changes.
Compare that to a real song from the real album.
On a more accessible note, Dadabots—which will continue to produce new albums on its Bandcamp site—also recorded the album Deep the Beatles, which was taken from the Beatles’ 1 compilation album (though be forewarned, it’s a rough listen if you’re expecting it to sound like, you know, the Beatles).
So, what does this mean for the future of music? Will AI eventually take over? Probably not, says one expert, especially when you consider people have been trying to create computer-generated music since the 1950s, perhaps most notably with Brian Eno in the 1990s.
“Nowadays everyone has a top quality digital camera in their phone, but this doesn’t mean that professional photographers have disappeared,” Valerio Velardo, the CEO of Melodrive and a self-described AI expert, composer, and conductor, told DW. “I think that it will be the same with AI. I completely doubt that the role of the composer will be harmed by this technology.”
H/T Digital Trends
Josh Katzowitz is a staff writer at the Daily Dot specializing in YouTube and boxing. His work has appeared in the New York Times, Wall Street Journal, Washington Post, and Los Angeles Times. A longtime sports writer, he's covered the NFL for CBSSports.com and boxing for Forbes. His work has been noted twice in the Best American Sports Writing book series.