Music Generated by Artificial Intelligence Hits Radio Sooner Than You Think – National | Globalnews.ca

And it works pretty well. In January 2021, Supertone introduced its Singing Voice Synthesis technology.

The big party trick was to introduce a new brand new song to Kim Kwang-seok, a Korean folk superstar who sold millions of records back home. Pretty cool considering he died in 1996.

Using artificial intelligence, Supertone’s SVS technology “learned” about 100 songs from 20 different singers to develop a style. Then it learned 10 songs from Kim. Put it all together and the AI ​​was able to create something that was more than a reasonable facsimile.

Why would HYBE be interested in such a technology? Because it’s the company behind some big K-pop acts, including BTS. Last year, the boy band shocked their worldwide fanbase by announcing that they were taking a break from music. Fair enough, given the insane ride they’ve been on for the past few years.

However, this created some serious problems. First, under South Korean law, every member of BTS is now required to complete compulsory military service (they were exempt from it due to artist and athlete exceptions).

This will keep BTS out of the limelight for at least 18 months. And since the seven members are of different ages, the times at which they are scheduled to begin their service will be staggered. BTS could be MIA for years. Not good for a $3.6 billion annual revenue company.

Could HYBE use Supertone to create new BTS footage while the boys are in the army? It seems possible.

From a purely capitalist point of view, this seems brilliant. No more spirited musicians who get drunk, get high, and end up doing stupid #MeToo things with fans. New music can be called on-demand, so you no longer have to wait for inspiration. Also, talk about music that is cheap to produce. No salaries, royalties, per diems, or other expenses eaten up by real people.

This begs the question: if the fake is indistinguishable from the real thing, will fans fall for it? Maybe.

Music synthesis by machine has been a dream of scientists for decades. In 1961, Max Matthews, a programmer at IBM, played around with a 7905 mainframe and made it sing. No computer had ever done that before.

Stanley Kubrick would later use this as the basis for a key scene 2001: A Space Odyssey in 1968.

What was cutting-edge research and science fiction is now very, very real. And while machines are not yet fully autonomous composers and performers, we are moving in that direction. For now, however, the focus is on AI-powered music creation software as a tool. Call it software-assisted composition.

In 2020, Grimes (the Canadian singer and ex of Elon Musk) collaborated with a startup called Endel to create a new piece of music she dubbed “AI Lullaby”. She created “stems” (short individual clips) of the music and vocals and left the rest to the software. Endel was also used to create music to help people fall asleep and driving music to help Mercedes-Benz drivers focus on the road.

Google is working on a system called AudioLM that can produce both natural-sounding speech/singing and music. All it takes is a few seconds of original audio and it will go from there. His piano pieces are smooth, flowing and somewhat nuanced. Also, no piano is required.

4:45
Tech Talk: AI technology takes over the voice of Darth Vader

Harmonai is a project by a company called Stability AI, which describes itself as “a community-driven organization releasing open-source generative audio tools to make music production more accessible and fun for everyone”. It also has a tool called Dance Diffusion (currently in beta) that can generate new, original short music clips based on its knowledge of a music catalogue. Some artists use the software as a food for thought to start a new composition.

Then there’s Amazon, which is working on software that can mimic the voices of the dead. Did grandma die? Give Alexa a few seconds of her voice and your device can talk to you like grandma for as long as you can stand it. I’m sure it’s just a few lines of code to get Grandma to sing your favorite Rage Against the Machine song.

Another area that sees potential is the industry that composes scores for film, television and commercials. This arena needs thousands upon thousands of new, short pieces of music every day, a need that is currently being met by human musicians. AI-generated production music will be cheaper, faster, and royalty-free.

In December 2015, a UK company called Jukedeck began offering users the ability to create five songs a month for free, and $7 for each track thereafter. Amper Music, an American company, followed in 2017, and Amadeus Code in 2019.

There’s also Germany’s Loudly (royalty-free production tracks), AVIA (“the artificial intelligence that composes emotional soundtrack music”), Infinite Album (AI-generated music for video games), and DAACI (“composes, arranges, orchestrates, and produces” original music.) And if you’re a podcaster who needs theme music, you might want to check out India’s Beatoven. There are dozens of other startups making real strides, including AI Music!, which is now owned by Apple.

Speaking of royalties, there’s an interesting debate going on. When a machine creates a new intellectual property, such as music, who owns the copyright to that work? The machine? The programmer? The manufacturer of the hardware/software? This still needs to be worked out in some areas.

There are those who believe that machines will never be able to achieve the emotions that music conveys. To really process the trauma of a DIVORCE, you have to go through one thing. On the other hand, it might only require a few extra lines of code.

—

Alan Cross is a broadcaster on Q107 and 102.1 the Edge and a commentator for Global News.

Subscribe now to Alan’s Ongoing History of New Music Podcast on Apple Podcast or Google Play