Non-native Listeners’ Recognition of High-Variability Speech Using PRESTO

• DOI: 10.3766/jaaa.25.9.9
• Publication Date: 2014-11-20T04:41:14Z
• Authors: Terrin N. Tamati; David B. Pisoni
• Keywords: Variation (astronomy)

Background: Natural variability in speech is a significant challenge to robust successful spoken word recognition. In everyday listening environments, listeners must quickly adapt and adjust multiple sources of both the signal environments. High-variability may be particularly difficult understand for non-native listeners, who have less experience with second language (L2) phonological system detailed knowledge sociolinguistic variation L2. Purpose: The purpose this study was investigate effects high-variability sentences on recognition explore underlying individual differences abilities listeners. Research Design: Participants completed two sentence tasks involving low-variability sentences. They also battery behavioral self-report questionnaires designed assess their indexical processing skills, vocabulary knowledge, several core neurocognitive abilities. Study Sample: Native speakers Mandarin (n = 25) living United States recruited from Indiana University community participated current study. A native comparison group consisted scores obtained English 21) taken an earlier Data Collection Analysis: Speech conditions assessed task using PRESTO (Perceptually Robust Sentence Test Open-Set) mixed 6-talker multitalker babble. HINT (Hearing Noise Test) Indexical skills were measured talker discrimination task, gender forced-choice regional dialect categorization task. Vocabulary WordFam familiarity test, executive functioning BRIEF-A (Behavioral Rating Inventory Executive Function – Adult Version) questionnaire. Scores compared tested previous examined differences. Results: Non-native keyword significantly lower than listeners’ tasks. Differences performance between larger HINT, although varied by signal-to-noise ratio. groups differed ability categorize talkers region origin knowledge. Individual accuracy babble at more favorable ratios found related subscales composite scores. However, not categorization, discrimination, or Conclusions: challenging Difficulty under testing lack L2, especially L2 information, among weaknesses affecting control life.