All of you that know me knows that I’m in love with music. So taking advantage from the class by Sergi Jordà in our posgraduate course in Information Visualization, this post is dedicated to briefly summarize the most common approaches used to visualize music. (Hopefully, future posts will be dedicated to more novel approaches…)
The most classical visualization of a sound is the line plot that reveals the sound’s signal. Shaped as a waveform, this visualization shows the physical phenomena that accours when we make a sound in what it’s called the time domain:
In this line plot, the x-axis represents time and the y-axis reveals the amplitude of the signal.
In order to better understand the sound that is being analyzed, the Fast Fourier Transform can be used to plot our sound in the frequency domain:
In this line plot the x-axis represents frequencies while the y-axis represents the sound level. In this case, the visualization helps us to discover the different frequencies that appear in a specific moment. The accumulation of line plots from the frequency domain using a third dimension that represents time generates the spectrum of the whole song:
The 3D spectrum of a song (Image from http://donrathjr.com/wp-content/uploads/2010/01/3D-SPectrum-of-Sound.jpg)
However, as usually happens with 3D visualizations, there is a lot of overlapping that prevents understanding the information in this picture. The following plot is usually used instead, which in visualization terms can be understood as a heatmap, where colors indicate the height of the waves:
This simple post is a small summary of the basic (and old!) concepts used to plot sounds/music.