A low-cost platform for voice monitoring

A low-cost platform is proposed in this paper that has been conceived to monitor the vocal activity of people that use the voice as a professional tool. Such a platform includes a wearable data-logger and a processing program that allows the vocal parameters to be extracted from the recorded signal. The data-logger is equipped with a contact microphone that is attached to the jugular notch of the person under monitoring, thus sensing the skin acceleration level due to the vibration of the vocal folds. The microphone output is conditioned through a custom circuitry and then sent to a cheap micro-controller-based board, which stores the raw samples onto a micro SD-card. The off-line processing provides an estimation of Sound Pressure Level (SPL), fundamental frequency (F0) and Time Dose (Dt), which are the parameters that seem most suitable for the identification of vocal disorders and the prevention of an improper use of the voice. For the estimated parameters, suitable calibration procedures are implemented and their effectiveness is shown through specifically conceived experimental tests. Experimental results are shown that refer to the calibration of the device and its normal use during monitoring interval of several hours. A comparison with a commercial device is also reported.