Acquisition of French Nasal Vowels ɛ̃, ã; e ɔ̃ by Brazilian Learners: Articulatory Aspects

Abstract: This work aims at investigating the acquisition of the nasal vowels ɛ̃, ã; and ɔ̃ of French by learners in an undergraduate course in Languages (Portuguese and French). In articulatory terms, this class of segments is characterized by a lowering of the velum, which makes a connection between nasal and oral tubes (SEARA, 2000; MEDEIROS e DEMOLIN, 2006; BARBOSA e MADUREIRA, 2015). Consequently, these vowels are characterized by acoustic aspects different from those found in oral vowels. With the free passage of air in the nasal tract, the first formant value (F1) tends to lower and the third formant value (F3) tends to rise (HAWKINS e STEVENS, 1985; DELVAUX, 2003). There is also a change in the width of spectral peaks - which is smaller for nasal vowels -, as well as the appearance of additional spectral peaks and a longer duration for these segments (MORAES e WETZELS, 1992; SOUSA, 1994). In order to develop this research, sessions of articulatory data collection have been made with three groups of informants: Group I - two learners of French as a Foreign Language (FFL) in different semesters (2nd and 8th) of the course; Group II - a native speaker of French; and Group III - a native speaker of Brazilian Portuguese. The articulatory data collection instrument consisted in a test involving the production of pseudowords in a carrier phrase. The sessions of data collection were made in an acoustic cabin of the Emergence of Oral Language Laboratory (LELO/UFPel). For the purpose of collection and analysis of articulatory data, the software Articulate Assistant Advanced (AAA, version 2.16.11) has been used. The analysis of the segments produced by both groups of native speakers has demonstrated that in order to distinguish the nasal segment from the oral one, the native speaker of French posteriorizes her tongue movements, whereas the speaker of Portuguese elevates them. Regarding the data produced by learners, it was possible to verify: (i) generalization of tongue gestures by the 2nd semester informant for the three French nasal vowels and (ii) accurate vocalic distinction by the 8th semester informant. Additionally, ultrasonography has shown to be a promising tool for research on nasality, evidencing significant statistic differences in tongue movements between oral and nasal segments.

摘要：本研究旨在探究本科语言类课程（葡萄牙语与法语方向）学习者对法语鼻元音ɛ̃、ã、ɔ̃的习得情况。从发音学视角来看，此类音段的典型特征为软腭下垂，使鼻腔与口腔声道形成连通（SEARA, 2000; MEDEIROS 与 DEMOLIN, 2006; BARBOSA 与 MADUREIRA, 2015）。正因如此，此类元音的声学特征与口腔元音存在显著差异。当气流可自由通过鼻腔声道时，第一共振峰（first formant, F1）数值通常会降低，而第三共振峰（third formant, F3）数值则趋于升高（HAWKINS 与 STEVENS, 1985; DELVAUX, 2003）。此外，其频谱峰值宽度会发生变化——鼻元音的频谱峰值宽度更小，同时还会出现额外的频谱峰值，且此类音段的时长更长（MORAES 与 WETZELS, 1992; SOUSA, 1994）。

为开展本研究，研究人员针对三组受试对象开展了发音数据采集实验：第一组为两名不同学期（第2学期与第8学期）的法语作为外语（French as a Foreign Language, FFL）学习者；第二组为法语母语者；第三组为巴西葡萄牙语母语者。发音数据采集工具为一项包含载体短语中假词产出任务的测试。数据采集环节在口语语言涌现实验室（LELO/UFPel）的隔音舱内完成。为采集与分析发音数据，本研究使用了Articulate Assistant Advanced（AAA，版本2.16.11）软件。

对两组母语者产出的音段进行分析后发现，为区分鼻化音段与口腔音段，法语母语者会将舌位后移，而葡萄牙语母语者则会将舌位抬高。针对学习者产出的数据，研究可验证两项结论：（i）第2学期受试对象针对三个法语鼻元音均存在舌部动作泛化的现象；（ii）第8学期受试对象能够准确区分各类元音。此外，超声成像被证明是鼻化现象研究的极具潜力的工具，其结果显示口腔音段与鼻化音段的舌部动作存在具有统计学意义的显著差异。