Skip to contents

Analysis Example

analysis_example.Rmd

Introduction

These analyses are the ones reported in paper titled “onsetsync: An R Package for Onset Synchrony Analysis” (submitted) by Tuomas Eerola and Martin Clayton. The sections and figures correspond to the manuscript.

Load libraries 

library(dplyr)
library(ggplot2)
devtools::install_github("tuomaseerola/onsetsync")
library(onsetsync)
packageVersion("onsetsync")
## [1] '0.5.1'

Read data 

data("CSS_IEMP", package = "onsetsync")
CSS_Song2 <- CSS_IEMP[[2]]
CSS_Song2 <- dplyr::select(CSS_Song2,Piece,Section,Clave,Bass,Guitar,Tres,SD,Cycle,Isochronous.SD.Time) # delete previous
print(knitr::kable(head(CSS_Song2),digits = 2,format = if(latex_format==TRUE){'latex'} else {'markdown'}))
Piece Section Clave Bass Guitar Tres SD Cycle Isochronous.SD.Time
Song_2 Son NA NA NA NA 1 1 5.04
Song_2 Son NA NA 5.28 NA 2 1 5.26
Song_2 Son NA NA 5.48 NA 3 1 5.48
Song_2 Son NA 5.71 5.71 5.73 4 1 5.71
Song_2 Son NA 5.93 5.94 5.92 5 1 5.93
Song_2 Son NA NA 6.15 6.14 6 1 6.15

Summarise onsets 

tab1 <- summarise_onsets(df = CSS_Song2, 
                         instr = c('Clave','Bass','Guitar','Tres','Isochronous.SD.Time'))
print(knitr::kable(tab1, digits = 1,format = if(latex_format==TRUE){'latex'} else {'markdown'}))
N Md M SD Min Max
Clave 486 666.4 703.6 173.9 192.0 1558.1
Bass 486 471.4 708.1 432.0 180.0 1985.2
Guitar 1401 223.6 244.5 91.4 175.1 1694.9
Tres 906 245.0 371.4 234.6 147.1 1986.5
Isochronous.SD.Time 1567 220.4 220.1 2.9 212.8 226.6

Figure 1 

print(plot_by_beat(df = CSS_Song2, 
                   instr = c('Bass','Clave','Guitar','Tres'), 
                   beat = 'SD', 
                   virtual='Isochronous.SD.Time',
                   pcols=2))

Analysis of Synchrony

Begin with a diagnostic plot.

Figure 2 

print(plot_by_beat(df = CSS_Song2,
                     instr = c('Guitar','Tres'),
                     beat = 'SD',
                     virtual = 'Isochronous.SD.Time', 
                     griddeviations = TRUE))

Synchrony between the instruments 

d1 <- sync_sample_paired(CSS_Song2, instr1 = "Guitar", instr2 = "Tres", beat = "SD")
dplyr::summarise(data.frame(d1), N = n(), Mean.ms = mean(asynch*1000), Sd.ms = sd(asynch*1000))
##     N  Mean.ms    Sd.ms
## 1 853 12.53126 26.74134

Figure 3 

inst <- c('Clave','Bass','Guitar','Tres') # Define instruments 
dn <- sync_execute_pairs(CSS_Song2,inst,beat = "SD")
print(plot_by_pair(dn))

Plain summaries of asynchrony

Reporting of several measures of absolute asyncronies.

d <- sync_sample_paired(CSS_Song2,'Clave','Bass',beat = 'SD')
print(summarise_sync(d))
##   Pairwise asynchronization Mean absolute asynchrony Mean pairwise asynchrony
## 1                  19.58636                  20.6644                 16.23288

Relative measure for two instruments vs other other instruments.

x<-sync_sample_paired_relative(df = CSS_Song2,instr = 'Bass',instr_ref = c('Guitar','Bass','Tres','Clave'),beat = 'SD')
print(x$`Mean pairwise asynchrony`)
## [1] -8.372744
x <- sync_sample_paired_relative(df = CSS_Song2,instr = 'Clave',instr_ref = c('Guitar','Bass','Tres','Clave'),beat = 'SD')
print(x$`Mean pairwise asynchrony`)
## [1] 5.381136

Asynchrony across beat subdivisions 

dn <- sync_execute_pairs(df = CSS_Song2, instruments = c("Guitar", "Tres"),beat = "SD")
print(summarise_sync_by_pair(dn, bybeat = TRUE))
##    Subdivision  N         M       SD        T   pval
## 1            1 48  5.034458 21.86708 1.595080 >0.999
## 2            2 34  9.983588 29.59437 1.967057  0.922
## 3            3 56  9.276143 32.68776 2.123617  0.611
## 4            4 76 17.359053 31.62904 4.784613 <0.001
## 5            5 52 10.107481 23.25047 3.134821  0.046
## 6            6 26  8.765923 27.32327 1.635881 >0.999
## 7            7 83  7.530265 25.72852 2.666456  0.148
## 8            8 70 14.508114 26.98793 4.497699 <0.001
## 9            9 44 17.526227 18.85664 6.165245 <0.001
## 10          10 41  5.288756 24.67663 1.372334 >0.999
## 11          11 29 13.804897 25.34219 2.933513  0.106
## 12          12 70 19.511314 29.21971 5.586755 <0.001
## 13          13 68  9.065206 20.76244 3.600425  0.010
## 14          14 17 14.309529 34.51096 1.709593 >0.999
## 15          15 68  7.444000 21.43886 2.863249  0.089
## 16          16 71 24.721338 26.68690 7.805540 <0.001

Synchrony across several performances

Using the built-in corpus of Cuban Salsa and Son preformances (CSS_IEMP).

corpus <- onsetsync::CSS_IEMP
D <- sync_sample_paired(corpus,'Guitar','Tres',n=0,beat='SD')
D <- D$asynch
D$asynch_abs <- abs(D$asynch)*1000
print(paste('Mean asynchrony across all performances is', round(mean(D$asynch)*1000,1),'ms'))
## [1] "Mean asynchrony across all performances is 17.3 ms"

Figure 4 

print(plot_by_dataset(D,'asynch_abs','name', box = TRUE) + ylab('Absolute Asynchrony (ms)'))

Synchrony with other variables 

CSS_Song2 <- CSS_Song2 %>%
  group_by(Cycle) %>%
  mutate(Tempo = 240/(max(Isochronous.SD.Time) - min(Isochronous.SD.Time))) %>%
  ungroup()

d2 <- sync_sample_paired(CSS_Song2,"Tres","Guitar", beat = 'Tempo')
d3 <- sync_sample_paired(CSS_Song2,"Tres","Guitar", beat = 'Isochronous.SD.Time')
tmp <- data.frame(asynch = d2$asynch*1000, Tempo = d2$beatL, Time = d3$beatL)
fig<-plot_by_var_time(tmp,var1 = "Time", var2="asynch",var3="Tempo",ylabel='Synchrony (ms)')
print(fig)