Monica McHenry1. 1. University of Houston, TX, USA. mmchenry@uh.edu
Abstract
PURPOSE: This study was designed to assess potential contributors to listener variability in judgments of intelligibility. METHOD: A total of 228 unfamiliar everyday listeners judged speech samples from 3 individuals with dysarthria. Samples were the single-word phonetic contrast test, the Sentence Intelligibility Test, an unpredictable sentence intelligibility test, and conversational speech. RESULTS: Across speakers, significant variability was found for all samples except the phonetic contrast test. Across tasks, significant variability was found for all speakers. There were no significant differences in age, gender, or education between the highest and lowest scoring listeners on the phonetic contrast test. CONCLUSIONS: These findings suggest that seemingly objective intelligibility tests are subject to a number of factors that affect scores.
PURPOSE: This study was designed to assess potential contributors to listener variability in judgments of intelligibility. METHOD: A total of 228 unfamiliar everyday listeners judged speech samples from 3 individuals with dysarthria. Samples were the single-word phonetic contrast test, the Sentence Intelligibility Test, an unpredictable sentence intelligibility test, and conversational speech. RESULTS: Across speakers, significant variability was found for all samples except the phonetic contrast test. Across tasks, significant variability was found for all speakers. There were no significant differences in age, gender, or education between the highest and lowest scoring listeners on the phonetic contrast test. CONCLUSIONS: These findings suggest that seemingly objective intelligibility tests are subject to a number of factors that affect scores.
Authors: Visar Berisha; Julie Liss; Steven Sandoval; Rene Utianski; Andreas Spanias Journal: Proc IEEE Int Conf Acoust Speech Signal Process Date: 2014-05