“…The second box in Figure 1 is a structured prediction algorithm for measurement of VOT (Sonderegger & Keshet, 2012; see Ryant, Yuan, & Liberman, 2013, for an alternative approach). Many standard approaches measure parameters based on pre-programmed rules developed in consultation with expert annotators (Boyce, Fell, MacAuslan, & Wilde, 2010; Hansen, Gray, & Kim, 2010; Prathosh, Ramakrishnan, & Ananthapadmanabha, 2014; Stouten & van Hamme, 2009). In contrast, the algorithm utilized here was designed to minimize the error in the predicted measurement and had a unique feature set.…”