ADULT-LIST EQUIVALENCY TESTING AND COMPARISON OF SPEECH RECOGNITION BETWEEN MALE AND FEMALE VOICE

Trimmis N (1), Markatos N (1), Malaperdas K (1), Papadas T (2)

(1) Department of Speech Pathology. Technological Educational Institute of Patras; (2) Department of Otolaryngology. University Hospital of Patras.

Background: Word recognition score tests should contain a number of equivalent word-lists. Male and female voices are sufficiently different to cause differences in word recognition scores.

Aims: The objective of this study was: firstly, to examine the list equivalency of the four phonemically balanced 50-bisyllabic-word lists, developed by Trimmis et al (2006), for speech recognition score testing, and secondly, to compare the performance-intensity function of each list between male and female voice.

Methods: One hundred adults (average age, 26.50 years; SD, 3.90) with normal hearing, divided in 2 groups (25 males and 25 females in each group) served as subjects. The lists with the male and the female voice were presented to group A and B, respectively at 11 intensity levels ranging from 0 to 50 dBHL in 5 dB increments.

Results: Statistical analysis of the results revealed no statistical significant differences among the four lists, either with male or female voice. In contrast, statistical significant differences were emerged on all four lists between male and female voice. In addition, steeper performance-intensity functions were obtained for the male recordings (M=6.1%/dBHL) than the female ones (M=5.8%/dBHL).

Conclusions: These findings suggest that all four lists were found to be equivalent for listeners with normal hearing whose native language is Modern Greek. Further research with hearing impaired subjects is required.

E-mail: trimmis@teipat.gr