Cough sound analysis to assess air quality in commercial weaner barns

Abstract The respiratory health and comfort of animals are closely associated with the air quality of their living environment. The aim of the present work is to identify the relationship between animal sounds and air quality of their living environment. Cough sounds of 84 weaners and four important aerial factors (i.e., air temperature, relative humidity, ammonia concentration, and dust concentration) were measured. The resulting power spectral density (PSD) showed a significant difference in cough sounds from weaners under different air qualities. Moreover, the developed recognition algorithm, using mel frequency cepstrum coefficients (MFCCs), principal component analysis (PCA), and support vector machine (SVM), provided an average recognition rate of 95% for sound samples collected in different pig houses. The results indicate that the cough sound analysis can provide qualitative information about the situation of the air quality in commercial livestock buildings.