Measurement of changes in arterial stiffness can be used as an early predictor of cardiovascular disease, the world's leading cause of death. Changes in arterial stiffness can be detected by measuring the pulse wave velocity (PWV) in arteries, since a wave propagates faster in a rigid medium and vice versa. Commercial devices measure the absolute value of PWV and compare it with clinically relevant population values, however, they do not analyze short-term changes in PWV. In this work, we developed a device that allows the measurement of PWV in peripheral arteries from the measurement of the photoplethysmography (PPG) wave at two sites of the arterial tree and its time lag, quantified by the pulse transit time (PTT). We performed PPG acquisitions in the brachial (forearm), radial (wrist), and digital (finger) artery, and obtained PTT values from 40 ms down to 3 ms, a difficulty due to the smoothness of the PPG signal. To address this challenge, we implemented three algorithms to calculate PTT: from the PPG signal peak, zero crossing, and maximum derivative. In all three cases, the PTT was stable and the PWV values were within the range reported in the literature for the different PPG measurement sites. The measurement of PPG in the proximal and distal phalanges is relevant due to the comfort for the individual. In this case, the maximum derivative method is the most robust. To assess the ability of the device to measure short-term changes in PWV, the hand was raised, generating a hydrostatic drop in blood pressure, and an expected drop in PWV was observed as arterial stiffness decreased. Preliminary results in one individual suggest that the developed device can be used to measure short-term local changes in PWV and relate it to changes in arterial stiffness.