What’s the Best Way to Pitch Shift and Time Stretch a Mashup?

Anh Dung Dinh – addinh@connect.ust.hk
Department of Computer Science and Engineering
The Hong Kong University of Science and Technology
Hong Kong SAR

Xinyang WU – xwuch@connect.ust.hk
Department of Computer Science and Engineering
The Hong Kong University of Science and Technology
Hong Kong SAR

Andrew Brian Horner – horner@cse.ust.hk
Department of Computer Science and Engineering
The Hong Kong University of Science and Technology
Hong Kong SAR

Popular version of 1pMU – Ideal tempo and pitch for two-source mashup
Presented at the 188th ASA Meeting
Read the abstract at https://eppro01.ativ.me/appinfo.php?page=Session&project=ASAICA25&id=3848002&server=eppro01.ativ.me

–The research described in this Acoustics Lay Language Paper may not have yet been peer reviewed–

Corey Blaz/Shutterstock.com

If you are a music enthusiast, chances are you have encountered mashups, a form of music remix combining multiple tracks together, on the Internet. DJs assemble a playlist of multiple popular songs with smooth transitions to spice up the radio station or club, and online artists layer tracks on top of each other to create a fresh take on existing songs.

To make a mashup that’s harmonically organized and pleasing, you need to consider the musical features of the original songs, including tempo – the speed at which the songs are played, and key – which musical notes are used. For example, let us combine the vocals and instrumental of these two songs:

“Twinkle Twinkle Little Star” melody rendered with vocal samples

“Vivacity” Kevin MacLeod (incompetech.com). Licensed under Creative Commons: By Attribution 4.0 License http://creativecommons.org/licenses/by/4.0/

There are different ways the songs could be modified to fit each other and combined. Some examples are shown here:

Our study aims to figure out which of the above examples, among others, would be rated by listeners as the best fit. We conducted a series of surveys to evaluate the preferences of over 70 listeners when presented with mashups of varying features. Our results are depicted in Figures 1 and 2 which show that most listeners preferred mashups with an average tempo and the original vocal pitch. More in-depth results are explored in our conference presentation and paper.

Figure 1: Average score of listener preference for different tempo variants in vocal-swap mashups. Higher score indicates more participants selected that option as the “most preferred” version of the mashup combining 2 songs. Overall, majority of listeners liked the mashups at average tempo of the two original tracks.

Figure 1: Average score of listener preference for different tempo variants in vocal-swap mashups. Higher score indicates more participants selected that option as the “most preferred” version of the mashup combining 2 songs. Overall, majority of listeners liked the mashups at average tempo of the two original tracks.

Figure 2: Average score of listener preference for different key variants, plotted as a function of the key differences between the 2 base songs. In most cases, the vocals’ original key is the most preferred version for the mashups.

Figure 2: Average score of listener preference for different key variants, plotted as a function of the key differences between the 2 base songs. In most cases, the vocals’ original key is the most preferred version for the mashups.

Our results will hopefully provide helpful insights for mashup artists to further enhance their compositions, as well as for automatic mashup creation algorithms to improve their output performance.