Speech-like monkeys' vocalizations are explained by four articulatory states compatible with Lieberman’s LDT

Fig. 1. The four articulatory states reachable by the vocal tract of monkeys. This panel is drawn with a superposition of acoustic data found by Boë et al. (2017) normalized in the maximal acoustic space and a selection of drawings of the macaque X-ray sagittal views extracted from Fitch et al. (2016). The 5 smooth tube-models having the same F1/F2 resonances as the vowel qualities /u ɔ ɑ ɨ æ/ found by Boë et al. (2017) in baboon’s vocalizations are paired with these drawings in order to show their high degree of similarity. The drawings are also placed similarly around the periphery of the convex hull found for the macaque but not exactly at the same F1/F2 frequencies. We see that the four articulatory states are obtained with elementary mouth and tongue maneuvers (see Hiimae et al., 1995) as following: (1) mouth closed/tongue retraction, (2,2’) mouth opened/tongue retraction, (3) mouth closed/tongue protraction and (4) mouth opened/tongue protraction. A unique tongue-soft palate constriction point is observed in (1) and (2,2’) forming a small pharyngeal cavity having a Helmholtz resonance between 1 and 1.5 kHz. The effective vocalic space (in red) is the subset of F1/F2 frequencies which are reachable with variations along the two axes and it is only a part of the wide vocalic space claimed by Boë et al. (2019) (in black). For simulations of the smooth tube models with Badin’s software (1984) (comparable to the lossy model), their length is fixed at 11.5 cm and the speed sound is set at 351 m/s without wall vibrations.
Hiiemae, K.M., Hayenga, S.M., Reese, A. (1995). Patterns of tongue and jaw movement in a cinefluorographic study of feeding in the macaque. Arch. of Oral Biol., 40 (3), 229-246.