Occupational voice : studying voice production and preventing voice problems with special emphasis on call-centre employees

This thesis deals with the occupational voice. In modern society, there is an increasing demand for oral communication in many professions. The label ‘occupational voice’ refers to those occupations where voice is the major tool. Employees in these occupations often suffer from voice symptoms to varying extents. In this thesis, the telephone customer service advisors are the target group of professional voice users. Telephone services constitute an example of expanding modern-day speech-related professional contexts. On the telephone, the speaker must rely solely on his/her voice without support from body language or written communication. Originally, initiation to gather data to be studied in this thesis came from the largest Finnish telecommunications operator Sonera (currently known as TeliaSonera Finland Oyj) in an effort to gain knowledge about the voice of their call-centre personnel. The first goal of this thesis was to investigate to what extent call-centre customer service advisors experience voice problems. The study showed that prolonged voice use can result in vocal symptoms even in the case of good environmental factors. Secondly, the aim was to find out how call-centre customer service advisors benefit from voice training. It was found that preventative voice training constitutes an efficient method to decrease voice symptoms. The third goal was to investigate how changes in voice production during a working day could be measured objectively using information embedded in the acoustic speech signal. Parameters were extracted both directly from the speech pressure signals and the glottal flows estimated by inverse filtering (IF). Only the fundamental frequency showed a statistically significant change during the working day. A possible explanation for the minor acoustical changes might be a good acoustical environment which did not make the employees raise their voice level. However, it was found that extracting acoustical parameters from running speech recorded in realistic work environments and communication situations is a prospective method to study voice production. To the best of the author’s knowledge, this was also the first time when continuous natural speech collected in a realistic occupational environment was used for IF. In addition, comparison between two IF methods showed consistent results hence encouraging IF to be a promising tool in the future research of occupational voice.