Evaluation of Voice Production by Acoustic Measurements in Mandibulectomy Patients

Abstract

The majority of studies on speech production have focused on articulation disorders; little is known about changes to the characteristics of voice. The purposes of this study were twofold. 1. Examine the acoustic parameters of computer-based acoustic analysis systems to determine which parameters are suitable to evaluate voice production in clinical settings. 2. Apply practical parameters to mandibulectomy patients to reveal the characteristics of their voice production. Thirty-eight normal subjects; 19 mandibulectomy patients associated with neck dissection and 6 mandibulectomy patients without neck dissection participated in this study. All voices were analyzed using a Computerized Speech Lab CSL system 4400 （Kay PENTAX, Lincoln Park, NJ, USA） and Multi Dimensional Voice Program （MDVP） software. The Noise-to-Harmonic Ratio （NHR） was determined to be a useful index for clinical assessment of voice quality. NHR values were significantly different between normal subjects and mandibulectomy patients with neck dissection, and between mandibulectomy patients with neck dissection and mandibulectomy patients without neck dissection. It was thus suggested that the voice quality of the mandibulectomy patients with neck dissection might have been influenced by their medical treatment. The cause of a patient’s speech disorder possibly originates from both articulation and voice disorders.