RECOGNITION OF SPOKEN WORDS

口语识别

• 批准号: 2126802
• 负责人: Mark A Pitt
• 金额: $ 8.46万
• 依托单位: OHIO STATE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 1993
• 资助国家: 美国
• 起止时间: 1993-02-01 至 1998-01-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/2126802
• 关键词: auditory stimulus; human subject; neural information processing; perception; psycholinguistics; semantics; speech; vocabulary

Theories of auditory word recognition posit multiple stages of mental representation to describe the transformation of speech into words; the final stage is the lexicon, whose operation and functions differ across theories. The proposed research examines three issues concerning the function of the lexicon in word recognition, testing predictions of classes of theories that are divided on each issue: (1) whether or not lexical knowledge is used to facilitate speech perception; (2) which portion of a spoken word first makes contact with its representation in the lexicon; (3) the process by which a lexical entry is selected that best matches the speech signal. Experimental manipulations are designed to reveal the operation of the lexicon. Stimuli will consist of words and sentences, and the acoustic, phonological, and semantic characteristics of words will be varied. Multiple experimental tasks, such as phoneme monitoring, phoneme identification, and cross-modal priming, will be used to probe processing at different stages in the recognition system and to provide converging evidence on the issues that will be studied. Specialized computer hardware, software, and audio equipment will control stimulus construction, presentation and data collection. The findings should contribute to an understanding of the role of the lexicon in word recognition.

听觉词再认的理论包括多个心理阶段 描述语音转化为单词的表示; 最后一个阶段是词汇，它的运作和功能在不同的语言中是不同的。 理论 拟议中的研究探讨了三个问题 词汇在单词识别中的作用，测试预测 在每个问题上划分的理论类别：（1）是否 词汇知识被用来促进言语感知;（2） 一个口语单词的一部分首先与它的代表接触 在词典中;（3）选择词条的过程 与语音信号最匹配。 实验操作的目的是揭示操作的 词典。 刺激将由单词和句子组成， 单词的声学、语音和语义特征将被 各种各样。 多个实验任务，如音素监测， 音素识别和跨通道启动，将被用来探测 在识别系统中的不同阶段进行处理，并提供 收集将要研究的问题的证据。 专门 计算机硬件、软件和音频设备将控制刺激 结构、列报和数据收集。 研究结果应该 有助于理解词汇在单词中的作用 识别.