Quantifying phonetic variation: Landmark labeling of imitated utterances
Quantifying phonetic variation: Landmark labeling of imitated utterances
复制标题
量化语音变化：模仿话语的地标标签
DOI：
--
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
S. Shattuck
中科院分区：
文献类型：
--
作者：
Jennifer Cole;S. Shattuck
众所周知，语音在不同的说话者和不同的环境中是高度可变的，听众会注意其中一些语音细节，因为它携带了丰富的上下文信息。在这一章中，我们介绍了一种从感知和产生的双重视角来研究语音变异的方法。我们分析了一系列的模仿听到的话语，其中要产生的语言对象是固定的句法，词汇和韵律，并采用了一种新的方法来量化语音变化使用声学地标（LMs）（史蒂文斯2002年）作为相关的语音对比方式功能。由10个美国英语母语者产生的模仿话语产生了3，500+辅音和元音LM，这些LM被标记并与词汇指定的LM和刺激中产生的LM进行比较。本研究主要观察到五个方面的结果：（1）在高度限制的模仿任务中，语言记忆的变化也会导致语音的弱化;（2）语言记忆的变化在不同的类别中是不对称的：元音语言记忆变化很少，而滑动语言记忆最容易发生变化;（3）LM缺失的某些模式在我们的数据中非常频繁，但没有语音变异模式在说话者之间或说话者内部的所有模仿实例中占主导地位：（4）根据LM识别的刺激中的语音简化形式不能可靠地模仿;（5）大约20%的词汇预测的语言记忆是以不同的结果产生的，无论是在说话者内部（跨重复）还是跨说话者。这些研究结果表明，并量化系统性的语音减少测量的LM。他们还表明，扬声器行使选择语音实施，偏离词汇目标和语音细节的听到的刺激。这些结果持有承诺的语音变异的研究中使用的模仿语音，并使用LM（和扩展其他功能线索）作为语音接地措施的语音生产的变化。
Speech is known to be highly variable across speakers and situations, and listeners pay attention to some of this phonetic detail for the rich contextual information it carries. In this chapter we introduce a method for investigating phonetic variation from the dual perspectives of perception and production. We analyse serial imitations of a heard utterance, where the linguistic object to be produced is fixed syntactically, lexically and prosodically, and employ a novel method for quantifying phonetic variation using acoustic landmarks (LMs) (Stevens 2002) as correlates of phonologically contrastive manner features. Imitated utterances produced by ten native speakers of American English resulted in 3,500+ consonant and vowel LMs, which were labelled and compared both to the lexically specified LMs, and to the LMs as produced in the stimulus. We report five main observations from this exploratory study: (1) Phonetic reduction due to variation in LM realization occurs even in the highly constrained imitation task; (2) variation is asymmetric across classes of LMs: Vowel LMs seldom vary, while glide LMs are most vulnerable; (3) certain patterns of LM deletion were very frequent in our data, but no pattern of phonetic variation prevailed over all imitated instances across or within speakers; (4) phonetically reduced forms in the stimulus, identified in terms of LMs, are not reliably imitated; (5) about 20% of lexically predicted LMs are produced with variable outcomes, both within speakers (across repetitions) and across speakers. These findings demonstrate and quantify systematicity in phonetic reduction as measured in terms of LMs. They also reveal that speakers exercise choice in phonetic implementation, deviating both from lexical targets and from the phonetic detail of the heard stimulus. These results hold promise for the use of imitated speech in the study of phonetic variation, and for the use of LMs (and by extension other feature cues) as a phonologically grounded measure of variation in speech production.
影响因子：
2.4
作者：
Stevens, KN
通讯作者：
Stevens, KN
影响因子：
8.2
作者：
Kraljic, Tanya;Samuel, Arthur G.;Brennan, Susan E.
通讯作者：
Brennan, Susan E.
影响因子：
2.4
作者：
Villacorta, Virgilio M.;Perkell, Joseph S.;Guenther, Frank H.
通讯作者：
Guenther, Frank H.