Can we use neuroscience and AI to predict successes? Maybe. – National – News Block

There’s a scene in the biopic. creation stories where Ewan Brenner, channeling Creation Records founder Alan McGee in a scene with his therapist, rants about the demands of finding The Next Big Thing in music. I’m spending millions on noises I have no idea if anyone will like it!” Welcome to the record business.

What makes a song a hit? Nobody knows. It is a mysterious organic process that no one has been able to unlock.

The president of a record company described things like this: “Running a record label is based on risk. We rely on creative types, musicians, to eventually provide us with songs that we hope the public will like it. A song can be objectively great, but if the public doesn’t bite, there’s no amount of money we can spend on marketing and promotion to do they like.”

This hasn’t stopped people from trying to find a way to accurately predict outcomes.

When rock’n’roll was still young, a couple of promoters got it into their heads that the process of writing hit songs could be summed up in one formulaic process. In 1959, Joe Mulhall and Paul Neff sent a questionnaire to 3,000 girls about their likes and dislikes when it came to music. Their thinking was that if they could incorporate as many positive data points as they could into a song, then they would be guaranteed a hit for their aspiring pop star, a 15-year-old American weightlifter named Johnny Restivo. When all the responses were collated, this song was the result.

The focus didn’t work. the way i am it only managed to reach number 80 on the pop charts.

There have been many attempts to find ways to predict hits, most often looking for people with “golden ears,” that amazing innate instinct certain people possess to hear success in something the public didn’t know they wanted. For example, in the early 1960s, the head of an American independent label began playing songs for his teenage daughter. He showed a real talent for predicting which of them would get it right (he had an 80 percent success rate), but it turned out to be beginner’s luck and his predictions failed after about 20 tries.

Meanwhile, the recording and radio industries built businesses around golden-eared people like Clive Davis (discoverer of Janis Joplin, Barry Manilow, Patti Smith, Whitney Houston, and many others); Mo Oistin (Fleetwood Mac, Prince, Red Hot Chili Peppers); Seymour Stein (The Ramones, Talking Heads, Madonna). Rosalie Trombley went from being a receptionist in CKLW/Windsor (The Big 8) to someone who had an uncanny ability to find hits. She not only convinced Elton John to launch benny and the jets as a single against all his reservations, but he chose hits from The Guess Who, Bob Seger, KISS and many others.

Others have taken different approaches. Weezer’s Rivers Cuomo uses a spreadsheet approach to songwriting, believing his next hit is hidden in the data. In 2003, Polyphonic HMI introduced Hit Song Science, a Barcelona-based artificial intelligence company, which used machine learning to analyze millions of data points obtained from Billboard hit songs dating back to 1955. The company he believed that he could unravel the underlying audio fundamentals of popular songs to not only explain their popularity, but also use that information to create new hits. while can have predicted the Grammy Award-winning success of Norah Jones’ debut album, Come with me (that’s up for debate), U2’s certified hits that ran on the project were rejected as flops.

Hit Song Science isn’t the only company trying to harness the predictive powers of AI. MusicXray, Bandmetrics, Mixcloud and a few others are also in this space. The potential payoff is huge. At least 100,000 new songs are uploaded to music streaming services every day, many of them junk. If anyone can come up with an idea to improve filtering algorithms to select only the best of the best, everyone from record labels to radio stations to streaming platforms will want to participate.

Perhaps this old data-driven approach is too limited. Welcome to the new field of music “neuroforecasting”. This is real pre-cog minority report things: Using the neural activity of a small group of people to predict future effects and behaviors of the mass population.

According to a report in Neuroscience.com, researchers in the US are augmenting AI machine learning with neural responses, i.e. brain waves, from living humans. Study subjects were fitted with ready-to-use physiological sensors, which collected brain activity associated with mood and energy levels. Different statistical approaches were applied to the data and machine learning was thrown into the mix and AI was applied to the neural responses recorded when real humans listened to a song.

The results were surprisingly good. The researchers claim a 97 percent accuracy rate when it came to predicting which songs would be hits. That’s much, much more than the 50 percent (a coin toss, really) derived from other, more traditional methods. To be fair, the test included only 33 people and their neural activity and involved 24 songs. But 97 percent is an outstanding success rate, if the technology actually works as advertised.

Will people with ears of gold and gut instincts for music get fired? I hope not. As good as any AI is, it can only mimic what it extracts from data and prompts. Only humans (for now) can get excited about something new and different.

However, as neuroprognosis is refined, it will have applications in endless areas of product testing and focus groups, for those companies and institutions that can afford it, of course. Music will be a great starting point. But where will it take us? We will have no choice but to wait and find out.

—

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

Subscribe to Alan’s ongoing new music podcast now on Apple Podcast or Google Play

&copy 2023 Corus Radio, a division of Corus Entertainment Inc.