基于Praat的听障生普通话语音偏误典型案例分析

A Typical Case Analysis of Mandarin Pronunciation Errors in Hearing-Impaired Students Based on Praat

摘要: 本研究旨在利用声学分析软件Praat对两名植入人工耳蜗的高中阶段听障生的普通话语音偏误进行案例分析。研究首先对朱川(1997)的《百字表》进行改进，设计了一份新的测试字表。随后，通过耳辨和Praat软件对两名被试的录音进行分析，将其语谱图与标准发音进行对比，从声母、韵母、声调三个维度系统地识别和归纳了语音偏误的特征。研究发现，听障生在舌尖前后音区分、清浊、送气、舌位控制、复元音动程以及鼻韵母等方面普遍存在发音困难。基于这些声学分析结果，研究提出了一套分级分层的语音感知训练方案，并结合“音书”App等工具，旨在为特殊教育领域的语音教学提供更精准、更具操作性的方法。

Abstract: This study aims to conduct a case analysis of the Mandarin pronunciation errors of two hearing-impaired senior high school students with cochlear implants, using the acoustic analysis software Praat. First, the study improves upon Zhu Chuan’s (1997) One Hundred Character List to design a new set of test characters. Subsequently, through auditory discrimination and analysis using Praat, the speech recordings of the two participants are examined. Their spectrograms are compared with standard pronunciations to systematically identify and summarize the characteristics of their pronunciation errors from three dimensions: initials, finals, and tones. The study finds that the hearing-impaired students commonly face difficulties in areas such as distinguishing between apical and retroflex consonants, voicing, aspiration, tongue position control, diphthong movement, and nasal finals. Based on these acoustic analysis results, the study proposes a tiered and graded speech perception training program. By incorporating tools such as the “Yinshu” App, the research aims to provide more precise and practical methods for speech instruction in the field of special education.