My science fair project consisted of determining whether adult voice experts, adult non-experts and minors could perceptually distinguish between smokers’ and non-smokers’ voices. It is my hypothesis that both medical experts and non-experts will be able to differentiate the voices of smokers from non-smokers, but experts might do better because of special training and experience. To the best of my knowledge, this hypothesis has not been previously studied.
The project was submitted to the Ethics Review Board of the Montreal General Hospital and approved. Patients, or study subjects, were recruited by a non-investigator from the Voice Laboratory and Otolaryngology clinics at the Montreal General Hospital. Subjects were interviewed to collect basic information such as whether they smoked, how much they smoked, and for how long they had smoked. To maximize evaluator ability to distinguish smokers from non-smokers, only heavy smokers were included. For the purposes of this study, smokers were defined as those who had smoked the equivalent of 1 package of cigarettes daily for 20 or more years; non-smokers had never smoked. Subjects were excluded for the following reasons: a history of vocal cord surgery, hoarseness, or recent laryngitis. Forty patients were recruited in total and consisted of 20 smokers and 20 non-smokers. The subjects were then asked to read a standard paragraph in the language of their choice (English or French) which was digitally recorded by a non-evaluator speech therapist. The recordings were then distributed as audio files to 3 groups consisting of three adult voice experts, two adult non-experts and two minors to determine, based on voice alone, the gender and smoking status of the subjects. Two voice doctors and a speech pathologist were the experts in this study. The results were then analyzed to determine the ability of the three groups to distinguish gender and the voices of smokers from non-smokers. At the end of the evaluations, data sheets were collected and entered into an Excel spreadsheet for analysis. The data were then looked at in the following ways: 1. the accuracy with which the different groups of evaluators distinguished smokers from non-smokers. 2. the accuracy with which the different groups determined gender, and 3. whether smoking status is more accurately determined in males or females.
® Computer
® Digital voice recorder
® Headphones
® CD