Acoustic analyses clarify voiced-voiceless distinction in tracheoesophageal speech

Abstract

In order to clarify the ability of the voice to achieve voiced voiceless distinction in [ce1]tracheoesophageal (TE) speech, acoustic cues such as closure duration, onset and offset of vibration during closure period and voice onset time (VOT), in conjunction with intraoral pressure, were analyzed in 40 TE speakers. Both closure period and VOT during [p] production were longer in TE speakers with high intelligibility compared with laryngeal speakers; during [b] production these parameters were similar between the two groups. TE speakers with high intelligibility and laryngeal speakers showed significant differences between [p] and [b] production in terms of both closure duration and VOT. TE speakers with low intelligibility of [b] had higher values of VOT during [b] production compared with those with high intelligibility. TE speakers with low intelligibility of [p] had lower values of VOT during [p] production compared with those with high intelligibility. It is concluded that these characteristic acoustic cues reflect voicing ability in TE speech.