Lifeboat Foundation

The complexity of the melodies of the most popular songs each year in the U.S.—according to the Billboard year-end singles charts—has decreased since 1950, a study published in Scientific Reports suggests.

Madeline Hamilton and Marcus Pearce analyzed the most prominent melodies (usually the vocal melody) of songs that reached the top five positions of the US Billboard year-end singles music charts each year between 1950 and 2022. They found that the complexity of song rhythms and pitch arrangements decreased over this period as the average number of notes played per second increased. They also identified two significant decreases in melodic complexity that occurred in 1975 and 2000, along with a smaller decrease in 1996.

The authors speculate that the melodic changes that occurred in 1975 could represent the rise of genres such as new wave, disco and stadium rock. Those occurring in 1996 and 2000 could represent the rise of hip-hop or the adoption of digital audio workstations, which enabled the repeated playing of audio loops, they add.

Randomly found on my hard drive.

Check the description to go a little faster.

Join us as we explore the different male personality types and the unique char…

Ever heard just a snippet of a song and instantly known what comes next? Or picked up the rhythm of a chorus after just a few notes? New research from the Center for Music in the Brain at Aarhus University and the Centre for Eudaimonia and Human Flourishing at the University of Oxford has uncovered what happens in our brain when we recognize and predict musical sequences.

Generative models have had remarkable success in various applications, from image and video generation to composing music and to language modeling. The problem is that we are lacking in theory, when it comes to the capabilities and limitations of generative models; understandably, this gap can seriously affect how we develop and use them down the line.

One of the main challenges has been the ability to effectively pick samples from complicated data patterns, especially given the limitations of traditional methods when dealing with the kind of high-dimensional and complex data commonly encountered in modern AI applications.

Now, a team of scientists led by Florent Krzakala and Lenka Zdeborová at EPFL has investigated the efficiency of modern neural network-based generative models. The study, published in PNAS, compares these contemporary methods against traditional sampling techniques, focusing on a specific class of probability distributions related to spin glasses and statistical inference problems.

Gentle shockwaves can help regenerate heart tissue.