The use of face masks as respiratory protective equipment (RPE) is considered key to maintaining the quality of life during emergency situations, long-term care and working conditions. Face masks can reduce the virus spread and bacterial infections as well as prevent the inhalation of industrial waste gases. Individuals who wear a facial mask over a prolonged time often reported uncomfortable feelings due to breathing resistance, heat, tightness, and overall discomfort. One of the main indicators used to quantify the level of discomfort induced by RPE is the respiratory rate (RR). In fact, RR can be directly associated with RPE-related unease since the presence of facial masks might intuitively modify the breathing pattern of the users. Unfortunately, still little is known about RR and its variability in response to wearing RPE. In the last year, the massive use of face masks due to COVID-19 pandemic fosters the development of sensors to measure RR once mounted into the medical mask. Among other, fiber Bragg grating sensors (FBGs) have gained growing attention since the intrinsic advantages of small size, lightweight, high metrological properties, and safety. In the present study, a single-use FFP2 surgical mask was instrumented by a soft sensor based on FBG to perform a longterm acquisition (i.e., 20 min) of the respiratory signal during ordinary work activities at the video terminal. The promising results confirmed the high accuracy of the proposed system in the estimation of RR with a maximum discrepancy of −0.69 breaths per minute and mean absolute percentage error of 2.88% when compared to a reference instrument. Moreover, no saturation of the sensor output occurred during the usage time.