Faculty of Arts and Sciences, Beijing Normal University, Zhuhai 519087, China.
School of Foreign Languages, Hunan University, Changsha 410082, China.
J Acoust Soc Am. 2024 Jul 1;156(1):655-671. doi: 10.1121/10.0028009.
The Kam language has experienced historical tonal splits, resulting in the development of a complex tonal system. However, there is still limited knowledge regarding the acoustic characteristics associated with aspiration-based tone splitting. This study aims to investigate the acoustic cues related to the tonal registers and laryngeal configurations in Donglei Kam, a dialect of Southern Kam. Sixteen native speakers of Donglei Kam participated, producing lexical tones. Statistical analyses were conducted to examine the acoustic distinctions between tonal registers, using measurements of voice onset time, spectral tilt, noise, and energy. The results indicated that Donglei Kam retained a two-way contrast of aspiration, albeit with a trend toward gradual loss. Additionally, a breathy voice was detected in the Ciyin tonal register, characterized by elevated spectral tilt values and spectral noise throughout the vowels. Moreover, machine learning classifiers effectively identified tonal registers using voice-quality data, suggesting that the phonation contrast between breathy and modal voice could contribute to the tonal split alongside pitch contrast. In summary, these findings enhance our understanding of the acoustic implementation of breathiness in Kam and offer valuable insights into the role of laryngeal contrast in tonal splits.
卡姆语经历了历史上的声调分裂，从而发展出了一个复杂的声调系统。然而，对于与送气为基础的声调分裂相关的声学特征，我们仍然知之甚少。本研究旨在探究与多雷卡姆语声调及声门配置相关的声学线索，多雷卡姆语是卡姆语的一种方言。十六位多雷卡姆语母语者参与了实验，他们发出了词汇声调。使用声时、频谱倾斜、噪声和能量等测量值，对声调寄存器之间的声学区别进行了统计分析。结果表明，多雷卡姆语保留了有送气与无送气的双向对立，尽管这种对立有逐渐消失的趋势。此外，在次低元音 Ciyin 中检测到了气声，其特征是整个元音的频谱倾斜值和频谱噪声都较高。此外，使用音质数据的机器学习分类器能够有效地识别声调寄存器，这表明呼吸声和常态声之间的发音对比可能与音高对比一起促成了声调分裂。总的来说，这些发现增进了我们对卡姆语中呼吸声的声学实现的理解，并为声门对比在声调分裂中的作用提供了有价值的见解。
