Long-term vocal parameters in semi-anechoic and reverberant rooms

The acoustic design of a workspace helps enriching speech and preventing talkers from forcing their voice. Voice disorders may cause absenteeism from work to recover, and many voice professionals such as teachers underestimate risks depending on the improper use of voice. Talkers tend to adjust their vocal output on the base of the vocal task, the distance to the listener and the sound environment. Given the same room acoustics, distance to the listener and type of speech, face-to-face communication implies interspeaker differences in the sound pressure level (SPL) of the voice signal at a short distance from the mouth, fundamental frequency (F0) and phonation time percentage (Dt%). The measurement chain and the characteristic of the sample are other causes of variability. This study investigates on changings in SPL, F0 and Dt% over 19 Italian mother-tongue subjects aging between 22 to 34 yrs, 11 female and 8 male, neither hearing nor visual impaired, producing continuous 5 minute-long speeches in the semi-anechoic and reverberant rooms of INRiM, in Turin (Italy). Talkers spoke freely aiming to transmit information on something they knew well to a listener seated 6 m far. The speech production was monitored through the contemporary use of a calibrated class 1 sound level meter (SLM) and two portable voice monitoring devices (VMD). The SLM, used as a reference measurement device, was placed at a fixed distance from the speaker’s mouth. The VMDs, based on two different contact microphones placed at the jugular notch of the speaker, were used to sense the skin vibrations induced by the vocal folds activity. A comparison between the use of different instrumental chains was assessed. Subjective surveys on vocal comfort completed the data acquisition with the aim of finding out correspondences between objective and subjective impressions.