Vocal sound imitations provide a new challenge for understanding the coupling between articulatory mechanisms and the resulting audio. In this study, the classification of three articulatory categories, phonation, supraglottal myoelastic vibrations, and turbulence, have been modeled from audio recordings. Two data sets were assembled, consisting of different vocal imitations by four professional imitators and four non-professional speakers in two different experiments. The audio data were manually annotated by two experienced phoneticians using a detailed articulatory description scheme. A separate set of audio features was developed specifically for each category using both time-domain and spectral methods. For all time-frequency transformations, and for some secondary processing, the recently developed Auditory Receptive Fields Toolbox was used. Three different machine learning methods were applied for predicting the final articulatory categories. The result with the best generalization was found using an ensemble of multilayer perceptrons. The cross-validated classification accuracy was 96.8% for phonation, 90.8% for supraglottal myoelastic vibrations, and 89.0% for turbulence using all the 84 developed features. A final feature reduction to 22 features yielded similar results.

Individuals may have a different body odor, when they are sick compared to healthy. In the non-human animal literature, olfactory cues have been shown to predict avoidance of sick individuals. We tested whether the mere experimental activation of the innate immune system in healthy human individuals can make an individuals' body odor be perceived as more aversive (intense, unpleasant, and disgusting). Following an endotoxin injection (lipopolysaccharide; 0.6 ng/kg) that creates a transient systemic inflammation, individuals smelled more unpleasant compared to a placebo group (saline injection). Behavioral and chemical analyses of the body odor samples suggest that the volatile components of samples from sick individuals changed qualitatively rather than quantitatively. Our findings support the hypothesis that odor cues of inflammation in axillary sweat are detectable just a few hours after experimental activation of the innate immune system. As such, they may trigger behavioral avoidance, hence constituting a first line of defense against pathogens of infected conspecifics.