A variety of different methods are available for measuring cough. In clinical practice and most clinical trials subjective reporting of cough is relied upon, using scoring systems or visual analog scores (VAS). Although these measures give an indication of patients' perceptions of the severity of the symptom, they may be unreliable because they are influenced by other factors such as mood, vigilance, and expectations. An objective measure of cough would therefore be a valuable tool. In the last decade advances in computer technology and the availability of portable digital sound recording devices have resulted in a resurgence of interest in developing ambulatory systems for recording cough. The ultimate goal is an automated detection system of use in the wide variety of conditions that cause cough. Multidisciplinary teams of researchers around the world are applying techniques such as neural networks, voice recognition models, and other signal processing techniques to this problem. The main challenge is achieving high sensitivity with good discrimination of noncough signals. For cough sound detection, this is confounded by both the variability of the acoustics of cough sounds within and between individuals and the amount and variety of speech sounds that must be discriminated. Significant progress is being made and it is likely that accurate automated objective monitoring systems will be available in the near future. These systems have the potential to change the way cough is measured in clinical practice and clinical trials, allowing a better understanding of the effect of existing and novel treatments on this troublesome symptom.