Development of quantitative evaluation system of hypernasality

异常共振所致的上鼻音是腭裂或上颌骨切除术后最常见、最严重的语音障碍之一。高鼻症的客观评估尚未建立，导致诊断和治疗上的障碍。本研究旨在开发一种客观、定量的方法来评估鼻腔共振，特别是鼻音过多。取得了以下研究成果：1.经过对设备的反复改进，开发了语音分析系统。该系统能立即显示带通滤波语音样本声学处理得到的三分之一倍频程频谱。2.利用该系统对腭裂患者、上颌切除患者和正常人的鼻腔共振进行了声学评估。1/3倍频程频谱分析显示，在被判定为高鼻音的发音中，第一和第二共振峰之间的功率水平增加，第二和第三区域的功率水平降低。3.多元回归分析显示，经验丰富的语音病理学家对过度鼻音的严重程度与该系统获得的定量评估之间存在高度相关。这一结果表明了该系统的实用性。

Hypernasality attributable to abnormal resonance is one of the most frequent, severe speech disorders in the patients with cleft palate or maxillary resection. The objective evaluation of hypernasality has not been established, resulting in impairment in the diagnosis and treatment. This study was carried out to develop an objective, quantitative method for evaluating nasal resonance, in particular, hypernasality. Following results were obtained.1.After repetition of improvement for equipments, an voice analyzing system was developed. The system immediately displayd one-third octave spectra which were obtained from acoustical processing of band-pass filtered speech samples.2.Nasal resonance was acoustically assessed by this system in cleft palate patients, maxillectomy patients and normal subjects. Analysis of one-third octave spectra revealed an increase in power level between the first and second formants, and a reduction in power level in second and third regions among utterances judged to be hypernasal.3.Multiple regression analysis revealed a high correlation between the perceived severity of hypernasality by well-experienced speech pathologists and quantitative evaluation obtained by this system. This result indicated usefulness of the system.