Search

The aim of this study was twofold: to develop a method for collecting voice data from natural working environments; and to study loading changes. Changes caused by loading were approached from two perspectives: the relationship between subjective complaints and objectively measured voice features, and the changes of objective variables during one working day. Subjects consisted of female teachers from primary and secondary schools. Six separate studies were carried out, the number of subjects for each one ranging from 3 to 33. The subjects were 33–59 years old. Recordings were made of their voices both during breaks and lessons with a portable DAT recorder. The recordings made during the breaks were collected on three days of one week, and four timeson each week. The speech samples of the lessons were taken from the first and last lesson of the day. Subjects filled in a questionnaire about their voice, which divided them into two groups: those with few complaints (the FC group) and those with many complaints (the MC group). In addition, three subjects performed a laboratory voice loading task. The speech sample recorded during breaks was a maximally sustained [a], from which fundamental frequency (F0), jitter and shimmer were analyzed. The lesson samples were analyzed for F0, its standard deviation (F0 SD), sound pressure level (SPL), its standard deviation (SPL SD), F0 time ( = active vibration time of the vocal folds) and long-time average spectrum (LTAS). The laboratory samples (a simple reading task) were measured for F0. Additionally, an index (F0 x F0 time/ 1000) for assessing voice loading is presented. The results showed that voice complaints correlated with objectively measured voice features: the more the complaints, the higher the F0 and SPL, the larger the F0 SD and the steeper the tilt of the LTAS, the latter indicating that the voices of the MC group had a tendency towards a hypofunctional voice usage. In addition, at the end of the working week the shimmer values of the MC group were lower than those of the FC group. During the working day, the F0 rose, F0 SD and SPL SD increased and the LTAS levelled out (changes towards a hyperfunctional voice usage). Some of these changes are caused by the classroom situation, some by the teachers themselves, and some (obviously the changes of the F0 and the LTAS) by physiological alterations. Interestingly, the main changes, which were the F0 rise and levelling of the spectrum, occurred in the FC-group. The changes possibly reflect a normal adaptation of the human body, and they may act as an alarm system for avoiding excessive strain and exhaustion. The voice loading index correlated with the voice complaints. The experiment showed that the used method is well suited for the measurement of voices in working places. The maximally sustained phonation proved to be a suitable voice task for the measurement of the voice in working places where subjects have to do a task independently. The results also showed that F0 yields different values in different conditions and with different voice tasks. This deserves attention when defining standards for acoustic variables.