In kleinschmidt/votcorpora: VOT production data from various sources

library(dplyr)
library(ggplot2)
devtools::load_all()

Size of each data set

Number of talkers

vot %>%
  group_by(source, subject) %>%
  summarise() %>%
  tally() %>%
  knitr::kable()

Tokens of each phoneme per talker

vot %>%
  group_by(source, phoneme, subject) %>%
  tally() %>%
  summarise_each(funs(mean, min, max, n_talker=length), n) %>%
  knitr::kable(digits=0)

Marginal distributions

ggplot(vot, aes(x=vot, color=phoneme)) +
  geom_density() +
  facet_wrap(~source)

ggplot(vot, aes(x=vot, color=phoneme, group=paste(phoneme,subject))) +
  stat_density(alpha=0.2, geom='line', position='identity') +
  facet_wrap(~source)