Perceptual evaluation and acoustic analysis of pneumatic artificial larynx

Abstract

Objective To investigate the perceptual and acoustic characteristics of the pneumatic artificial larynx (PAL) and evaluate its speech ability and clinical value. Study Design Prospective study. Setting The study was conducted in the Voice Lab, Department of Otorhinolaryngology, The First Affiliated Hospital of Nanjing Medical University. Subjects and Methods Forty-six laryngectomy patients using the PAL were rated for intelligibility and fluency of speech. The voice signals of sustained vowel /a/ for 40 healthy controls and 42 successful patients using the PAL were measured by a computer system. The acoustic parameters and sound spectrographs were analyzed and compared between the two groups. Results Forty-two of 46 patients using the PAL (91.3%) acquired successful speech capability. The intelligibility scores of 42 successful PAL speakers ranged from 71 to 95 percent, and the intelligibility range of four unsuccessful speakers was 30 to 50 percent. The fluency was judged as good or excellent in 42 successful patients, and poor or fair in four unsuccessful patients. There was no significant difference in average fundamental frequency, maximum intensity, jitter, shimmer, and normalized noise energy (NNE) between 42 successful PAL speakers and 40 healthy controls, while the maximum phonation time (MPT) of PAL speakers was slightly lower than that of the controls. The sound spectrographs of the patients using the PAL approximated those of the healthy controls. Conclusion The PAL has the advantage of a high percentage of successful vocal rehabilitation. PAL speech is fluent and intelligible. The acoustic characteristics of the PAL are similar to those of a normal voice.