Mangle Positive Timing Function — mangle_positive

This function adds random jitter to a time column in a midi_df and ensures all times are positive. It samples jitter values from a uniform distribution, and then adds these to the original time series. Any negative times resulting from the jitter are set to 0.

Usage

mangle_positive_timing(x, ticks_amount = 5)

Arguments

x: A numeric vector representing time points.
ticks_amount: The range of the uniform distribution from which to sample jitter. Default is 5.

Value

A numeric vector with the same length as x, representing the 'mangled' time series.

Details

This function can be used with dplyr to modify a midi_df.

The result is similar to "humanizing" the timing. Depending on the amount of jitter, small random ticks are added to all timing messages. This function does not jitter any simulatenous messages, so chords do not become spread out as triads.

Examples

if (FALSE) {
mario <- midi_to_object("all_overworld.mid") #import midi
list2env(mario, .GlobalEnv) #send object to global environment

midi_df %>%
mutate(time = case_when(
  meta == FALSE ~ mangle_positive_timing(time, ticks_amount = 5),
  meta == TRUE ~ time)
  ) -> copy_df

# update miditapyr df
miditapyr_object$midi_frame_unnested$update_unnested_mf(copy_df)

# write midi file to disk
miditapyr_object$write_file("jitter_mario.mid")
}