For each recording, each identity space R script determined the duration of any one tune or voice type (total duration / number of types) if the distribution of total music seconds in the 'super corpus' were uniform.
If you are using the R scripts to analyze your own data, then you would use the R scripts to generate your own versions of these file. You do not need to use the files shared in this component, because they are specific to our music data.
If you are interested in re-generating the supplemental results reported here for Mendoza & Fausey: The feature space of music in everyday infancy, then you do need to use the files shared in this component. These files are the simulated data of the individual durations of tunes or voices across all recordings if each recording's 'super corpus' distribution was uniform. These files were read in to each R script to produce the supplemental results we report here.
