Add a more texty example to the workflow

[mdingemanse]

...with the caveat that label scaling is not veridical to utterance duration

E.g. this:

  ggplot(aes(x = line_end, y = line_participant)) +
  ggtitle("Relative frequency of elements within turns") +
  scale_y_reverse() + # we reverse the axis because lines run top to bottom
  geom_label(aes(
    begin = line_begin-800,
    end = line_end-800,
    label = utterance)) +
  xlab("Time (ms)") +
  ylab("") +
  theme_turnPlot()

[mdingemanse]

Very hacky, so more work needed, but here's what I'm calling a fish bone plot — a fairly readable way of looking at the temporal structure of a conversation. Code in elpaco-escience/playground@2a1689e (direct link).

[mdingemanse]

Or perhaps rework this one using clean talkr code (original in playground/continuers.R)

[mdingemanse]

Another idea is to take an example from Ruhleman and work it out in talkr