Cough sound analysis and objective correlation with spirometry and clinical diagnosis

Toward Mitigating Spreading of Coronavirus via Mobile Devices

“…• Mel-frequency cepstral coefficients (MFCC): MFCC is one of the most broadly used features for speech classification [13]. MFCC makes learning of an audio signal easy, and the total MFC coefficients range from 0 to 39, which makes a complete power spectrum of a signal on a Mel scale frequency.…”

“…We consider a total of 20 coefficients to achieve good performance. The frequencies are converted to Mel scale using the equations mentioned in [13]. • Variation rate: It measures the variation coefficient of the short-term energy [13], which is calculated by adding the squared absolute values of amplitudes normalized by the length of the frame.…”

“…The frequencies are converted to Mel scale using the equations mentioned in [13]. • Variation rate: It measures the variation coefficient of the short-term energy [13], which is calculated by adding the squared absolute values of amplitudes normalized by the length of the frame. • Zero crossing rate (ZCR): ZCR measures the sign changes of the amplitude values of each frame [13].…”

Automatic Detection of COVID-19 Based on Short-Duration Acoustic Smartphone Speech Analysis

“…Over the last decade, smartphone technology has shown promise as a convenient biosensor to help track many different categories of illnesses (e.g., cardiovascular, mental, neurological, respiratory) [ 5 – 10 ]. Further, using audio recordings collected via smartphone devices, biomedical studies [ 11 – 15 ] have investigated a variety of acoustic feature types and machine learning techniques to help automatically detect respiratory illnesses. For example, glottal speech features (e.g., glottal-to-noise excitation) have been explored in respiratory disease detection studies [ 11 ] to measure differences in fundamental frequency (F0), excitation cycle, and vocal tract airflow.…”

“…Suprasegmental prosodic-based speech features, such as pitch, formant frequencies, and loudness, have also been investigated for COPD/asthma-related illness detection [ 13 , 14 ]. Perhaps, the most commonly used speech features for automatic respiratory disease detection are spectral (e.g., cepstral) derived from the short-term power spectrum of the speech signal [ 5 , 14 , 15 ].…”