The link between speaker-specific accuracy in speech production and listener-specific acuity in speech perception: From subphonemic cue-weighting to prosodic categories

特定于说话者的语音生成准确性与特定于听者的语音感知敏锐度之间的联系：从亚音素线索加权到韵律类别

• 批准号: RGPIN-2018-06518
• 负责人: Pape, Daniel
• 金额: $ 1.75万
• 依托单位: McMaster University
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=713801
• 关键词: link; between; speaker; specific; accuracy

Speech production is concerned with the articulatory mechanisms to produce an acoustic speech output, whereas speech perception examines how differences in articulatory settings and acoustic variation contribute to a robust perceptual construct. However, currently there is sparse research on the direct connection between the articulatory accuracy of a speaker on the one hand and the acuity of speech perception of a listener, on the other hand. A number of interesting research questions arise from a closer examination of this production-perception link: (1) can an accuracy/acuity link be found for individual speakers/listeners as well, i.e. is a speaker that produces more accurate and less variable speech production targets also coincides also a listener with improved speech perception skills? And vice versa, (2) are listeners with well-defined and narrow perception boundaries also speakers with less variable and more precise places of articulation? And if examining differences between speakers/listeners from an individual point of view, (3) do certain individual personality traits and spectrum differences influence production accuracy and/or the perceptual acuity to distinguish speech sounds? Finally, if seen from a cross-population or cross-linguistic perspective, (4) can we define certain groups of speakers/listeners who would use only certain acoustic cues or a combination of different cues in their production and perception strategies, or is there rather a single major strategy across different populations/languages to achieve a certain production/perception contrast? The overarching goal of this project is to elucidate how speech production and speech perception in different languages and different populations are linked through their boundary conditions, namely (1) the articulatory requirements to produce speech and (2) the acoustic consequences of these articulatory requirements. This goal will be reached by examining different acoustic cues found within speech sounds and contrast these findings with cues that span several speech segments like syllables or words to elicit if (1) increased accuracy in the production of speech also leads to increased perceptual acuity and (2) what role acoustic variance plays for the generation of robust perceptual contrasts, in other words: what forms a phonetic/phonological category? This work will advance our understanding how speech production and perception are linked, i.e. which configuration of acoustic cues and/or articulatory settings results in an optimal perceptual output. Potential applications of this work are ultrasound videos with optimized articulatory and/or acoustic settings that help Speech-Language therapists to find the best therapy approach for a given speech sound disorder. Furthermore, the output of this program will help optimizing cochlear implants and thus benefit hearing-impaired patients.

言语产生与产生声学语音输出的发音机制有关，而言语知觉则研究发音环境的差异和声学变异如何促成强健的知觉结构。然而，目前关于说话人的发音准确度与听话人的言语感知敏锐度之间的直接联系的研究很少。 对这一产出-感知联系的更仔细研究产生了一些有趣的研究问题：(1)是否也能找到单个说话者/听者的准确度/敏锐度联系，即，产生更准确和更少变化的言语产出目标的说话者是否也与言语感知技能得到改善的听者相吻合？反之亦然，(2)知觉边界明确且狭窄的听者是否也具有较少变化且发音位置较精确的说话者？如果从个人的角度考察说话人/听话人之间的差异，(3)某些个性特征和频谱差异是否会影响发音的准确性和/或区分语音的感知敏锐度？最后，如果从跨人群或跨语言的角度来看，(4)我们能否定义特定的说话者/听者群体，他们在产生和感知策略中只使用特定的声音提示或不同提示的组合，或者是否存在跨不同群体/语言的相当单一的主要策略，以实现特定的产生/感知对比？ 这个项目的首要目标是阐明不同语言和不同人群的言语产生和言语感知是如何通过它们的边界条件联系在一起的，即(1)产生言语的发音要求和(2)这些发音要求的声学后果。这一目标将通过检查语音中发现的不同声学线索来实现，并将这些发现与跨越音节或单词等几个语音片段的线索进行比较，以得出以下结论：(1)语音产生的准确性增加也会导致知觉敏锐度增加，以及(2)声学差异在产生强健的知觉对比中起到什么作用，换句话说：什么构成语音/音系类别？ 这项工作将促进我们对言语产生和知觉是如何联系在一起的理解，即哪种声学提示和/或发音设置的配置会产生最佳的知觉输出。这项工作的潜在应用是具有优化发音和/或声学设置的超声视频，帮助言语语言治疗师为给定的言语声音障碍找到最佳治疗方法。此外，该程序的输出将有助于优化人工耳蜗植入，从而使听力受损患者受益。