The Beatles' music underwent a dramatic transformation in just a few years, but until now there was no scientific way to measure this progression

From the initial rock and roll to its leaks with psychedelia, The Beatles' music has undergone a constant evolution.

In addition to being one of the most influential groups in history, they are also recognized for their openness to experimentation in lyrics, melodies and the introduction of new instruments in classical styles.

Although this transformation is clear to any fan who has followed the group's trajectory, there was no scientific way to measure progression. Until now.

A team of computer scientists from Lawrence University of Technology has developed an algorithm that analyzes and compares musical styles, and they have done so on the musical progression of The Beatles.

The study, published in the Pattern Recognition Letters, scientifically demonstrates that The Beatles' music structure progressively changes from one album to another.

And so this algorithm works, first convert each song into a spectrogram (a visual representation of the audio content).

The algorithms convert each spectrogram into a set of almost 3,000 numerical descriptors that reflect visual aspects such as textures, shapes and the statistical distribution of pixels.

The researchers tested the algorithm with 11 songs from each of the 13 studio albums of The Beatles. As a result, artificial intelligence quantified the similarities between each set of songs and was able to perceive the difference between those that were part of an album and those that were not. But he was also able to find out what songs had been composed before others.

After this initial analysis, the algorithm chronologically ordered the studio albums and succeeded in order. This means that he concluded that the songs on the first album, Please, Please Me, looked more like the group of songs on the second album, With the Beatles, and less like the songs on the last album they recorded, Abbey Road. In fact, the algorithm was able to identify that Let it be songs (the last album they released for sale) had been recorded before those of Abbey Road (the last record they recorded).

"It is worth studying what makes The Beatles music so distinctive, and computer science and big data can help" - the developers comment.

The developers of this algorithm also analyzed the albums of other bands such as Queen, U2 and ABBA. In these cases, the algorithm was able to deduce the chronological order of the albums of other groups by analyzing only the audio data, without the spectrogram images.