Abstract
This study addressed the influence of clear speech on intelligibility and voice onset time for word-initial stops produced by electrolaryngeal speakers. Eighteen consonant-vowel-consonant English words containing /p/, /t/, /k/, /b/, /d/, and /g/ in word-initial position were spoken by 10 laryngectomees in habitual speech and clear speech conditions. Twelve, naive listeners transcribed a total of 4,320 words across habitual speech and clear speech conditions, of which 720 perceptual ratings (containing word-initial stops) were analyzed. Results indicate that electrolaryngeal speakers produced voiced word-initial stops with 25 milliseconds of longer voice onset time in clear speech compared to habitual speech and voiceless word-initial stops with 29 milliseconds of longer voice onset time in clear speech compared to habitual speech. The intelligibility of word-initial stops improved 3.8% during clear speech, while listeners omitted 2.22% more word-initial stops when electrolaryngeal speakers used habitual speech. Repeated measures analysis of variance revealed a significant effect of speaking condition on voice onset time, but not for the identification of word-initial stops. Collectively, these findings provide initial evidence that volitional attempts to make electrolaryngeal speech clearer via clear speech do not necessarily facilitate improved intelligibility for electrolaryngeal speakers in the presence of longer voice onset time durations.