Figure 3.

Workflow for cough audio processing. For segmentation phase, predefined thresholds can be chosen based on the physiology of cough sounds. The raw recorded sounds contains a lot of silent fragments (with low intensity) and background noise. Therefore, silence removal phase is required for saving storage space. For signal quality check, an estimate of the signal-to-noise ratio (SNR) can be computed by taking the ratio of the power of the cough part of the signal to the ratio of the rest of the signal.