With the rise of COVID-19, remote monitoring tools that can monitor or possibly identify respiratory conditions would be very beneficial. One of the most common symptoms of respiratory disease is cough, whose characteristics can provide medical insight to health care professionals when identifying the underlying cause (e.g. virus, bacteria, or acute/chronic condition) [13]. Cough characteristics are typically explored by a physician during an appointment by 1) asking the patient to describe the characteristics of the cough, e.g. frequency and productivity, 2) listening to involuntary coughs during the encounter, 3) listening to the lungs of the patient during deep breathing and 4) potentially asking the patient to produce a voluntary cough. However, voluntary coughs may have different characteristics when compared to an involuntary cough [14]. Self-report measures are well known to be inconsistent, which is especially true in the older adult population.The automatic characterization of cough based on some of the current medical ontology (wet cough, dry cough, whooping cough) may prove beneficial as an additional source of information for physicians when making medical decisions. Furthermore, characterization of specific underlying condition associated with a particular cough (e.g. COVID-19) may have implication in disease identification, monitoring and contact tracing in public areas.Studies have indicated that cough caused by COVID-19 is likely to have distinct latent features. In [15], the authors analyzed the pathomorphological changes caused by the COVID-19 in the respiratory system from the studies examining Xrays and CT scans of COVID-19 patients. Their study also included the autopsy report studies of deceased patients. Their findings suggest that cough sound signatures with COVID-19 are likely to have some idiosyncrasies stemming from the distinct underlying pathomorphological alterations. In an-Abstract-Remote monitoring and measurement are valuable tools for medical applications and they are particularly important in the context of pandemic outbreaks, like the current COVID-19. This paper presents an analysis of sound measurements of cough events from the point of view of their predictive content with respect to identification of different types of cough, including positive COVID-19 cases. The data consisted of a collection of audio samples collected from different sources including dry, wet, whooping and COVID-19 coughs. Unsupervised and supervised machine learning techniques were used to reveal the underlying structure of the data, described by dissimilarity spaces constructed from pair-wise dynamic time warping measures derived from the original sound measurements. Intrinsic dimensionality, nonlinear mappings to low-dimensional spaces and visual cluster assessment techniques allowed a representation of the cough types distribution. Supervised classification t echniques w ere u sed to obtain models identifying cough classes and high performance classifiers w ere o btained for m ost o f t hem, i ncludi...