Table 1. Basic comparison of Audiophile and Cliffhanger turn models.

Speakers’ mean and median turn durations were four to five times greater for the cliffhanger turn model compared to those of Audiophile. This suggests more broadly that analytic decisions about transcript segmentation will play a key role in the empirical investigation of speaking duration, an understudied topic.

Model	Turn duration		Number of words		Average turns
	Mean	Median	Mean	Median
Audiophile	2.22	0.92	6.40	2	440.70
Cliffhanger	8.52	5.81	17.81	9	159.41