Using speech as an index of alcohol intoxication
Chris Martin, Moshe Yuchtman · The Journal of the Acoustical Society of America · 1986
Although laypersons and low-enforcement officers routinely use perceived speech quality as an index of alcohol intoxication, there has been little research on the accuracy and limitations of this ability. In a recent study, we found that listeners could reliably discriminate between sentences produced by talkers speaking under sober and intoxicated conditions and that experienced listeners performed significantly better than naive listeners [Pisoni, Hathaway, and Yuchtman, SRL-PR ♯11, 109–172 (1985)]. In the present study, we compared the performance of college students and law-enforcement officers who were instructed to determine whether recorded sentences were produced under intoxication. In addition, digital signal processing techniques were used in an attempt to define changes in the acoustic-phonetic structure of speech under intoxication. The implications of this study for the role of speech as a reliable index of intoxication will be discussed. [Work supported by General Motors Research Laboratories.]