Speech normalization across speaker, sex and accent variation is handled similarly by listeners of different language backgrounds

This study assessed the influence of language background in speech normalization by examining non-native vowel categorization across speaker, sex and accent variation. Mandarin-English bilinguals, Australian English bilinguals and monolinguals categorized /ɪ/ and /ɛ/ produced by a female Dutch speaker, and were then tested with the same vowels produced by speakers of the same or different sex and/or accent. Listeners categorized the vowels regardless of speaker and sex variation, but showed lower accuracy when vowels were produced by speakers of different accent or accent and sex. Findings suggest that listeners normalize speaker and sex variation automatically, while accent variation requires contextualization