Computer scientists at Lawrence Technological University have created an artificial intelligence capable of studying how the Beatles changed their music over the years.
The technology was developed by Assistant Professor Lior Shamir and graduate Joe George, who used an algorithm they developed for technology used to study how whales vocally communicated, according to Phys.org.
Each Beatles song was converted into a spectrogram, which is a visual representation of the audio content. Each music spectrogram was then turned into a set of about 3,000 numeric descriptors that reflect texture, shapes, the pixels' statistical distribution and other aspects. The scientists then used statistical and recognition methods to find similarities between different songs.
11 songs from each of the Beatles' 13 studio albums released in the U.K. were studied, and the similarities between each other were quantified and compared to similarities between the albums, Science Daily reported. The way the albums were automatically placed matched up with the chronological order of each album's recording. The team started with "Please, Please Me," the band's first album, and followed it with "With the Beatles," "Beatles for Sale" and "A Hard Day's Night." The algorithm found that the songs from "Please, Please Me" were most similar to the songs from the second album. "With the Beatles," and least similar to "Abbey Road," the band's last recorded album.
The other albums were also placed in chronological order as well. While "Let it Be" was the last album that the Beatles released, its songs were found to have been recorded earlier than the songs on "Abbey Road."
"People who are not Beatles fans normally can't tell 'Help!' was recorded before 'Rubber Soul,' but the algorithm can," Shamir said. "This experiment demonstrates that artificial intelligence can identify the changes and progression in music styles by 'listening' to popular music albums in a completely new way."
Today's algorithms are capable of searching, browsing, and organizing huge databases of music, as well as finding songs that match an individual listener's music preferences, Phys.org reported. Shamir believes this technology will have a huge impact on the legacy of the Beatles.
"The baby boomers loved the music of the Beatles. I love the Beatles, and now my daughters and their friends love the Beatles. Their music will live on for a very long time," Shamir said. "It is worthwhile to study what makes their music so distinctive, and computer science and big data can help."
Shamir and George's research was published in the August issue of the journal Pattern Recognition Letters.
