In music, a mash-up refers to the act of combining several songs to create a new, different-sounding track.
The art of the mash-up originated in the late 1960s, under the impetus of the American singer-songwriter, Harry Nilsson. He remixed The Beatles’ “You Can’t Do That” in 1967, incorporating a dozen other Fab Four songs.
Since then, mash-ups have become a DJ specialty. But they could soon face competition from the latest invention of Xinyang Wu, a researcher at the Hong Kong University of Science and Technology.
The researcher has developed software that uses artificial intelligence to create mash-ups from the drum tracks of one song and the vocals and instruments of another.
In detail, it isolates extracts from each of the two tracks to be combined and selects those that are the most dynamic. The algorithms it employs adjust the tempo of the various instrumental tracks and add the drum beat at just the right moment to achieve “maximum effect,” explains a news release.
This technique is similar to that used by DJs and other music professionals to create mash-ups that are sure to get the crowd dancing.
“Imagine trying to make a gourmet meal with only a microwave – that’s sort of what automated mashup software is up against compared to a pro chef, or in this case, a professional music composer.
“These pros can get their hands on the original ingredients of a song – the separate vocals, drums, and instruments, known as stems – which lets them mix and match with precision,” Xinyang Wu explains in a statement.
This software, which was presented publicly at the 2023 edition of the Acoustics festival, has so far created mash-ups based on drum tracks.
However, Xinyang Wu is hoping to test it on other instrumental tracks with the aim of developing technology that could one day be used by all kinds of music fans.
“Our ultimate goal is creating an app where users can pick any two songs and choose how to mash them up – whether it’s switching out the drums, bass, instrumentals, or everything together with the other song’s vocals,” the researcher concludes.