A percepção das vogais do inglês norte-americano por falantes de inglês como le

• Autor(a) principal: Sacchi, Aline Cristina
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da PUC_SP
• Texto Completo: https://tede2.pucsp.br/handle/handle/21591
• topic: Fonética; Fonética acústica; Língua inglesa – Vogais; Percepção da fala; Phonetics; Phonetics, Acoustic; English language – Vowels; Speech perception; CNPQ::LINGUISTICA, LETRAS E ARTES::LINGUISTICA::LINGUISTICA APLICADA

This dissertation aims to study how bilinguals perceive English vowels. The subjects are students from a Canadian school in Sao Paulo who speak both English and Brazilian Portuguese fluently. The English language will be considered a foreign language (FL) in this context, because the subjects learn and use the language within a country in which English is not the first language. In the process of learning a new language some differences in terms of (FL) foreign language sounds may not be perceived due to interference of the first language (L1). Assimilation of L2 to L1 sounds may occur as the Speech Learning Model states, which explains the importance of formal teaching to train students to improve their fluency and intelligibility. The goal of this project is to investigate how bilingual speakers perceive English vowels using perception identification and discrimination tests. The judges are 27 students who study at a Canadian school in Sao Paulo, and whose L1 is Brazilian Portuguese and FL is English, and their average age is 9 years old. A Canadian native speaker of English at the age of 50 recorded the stimuli for the perception test containing the English vowels analyzed in this study. The hypothesis is that the subjects will not be able to discriminate between some sounds contrasts and will associate different sound pairs to only one Portuguese sound (Speech Learning Model, Flege). We conducted an experiment and analyzed the stimuli acoustic-articulatory characteristics based on acoustic inspection, formants and durations of the vowels were measured. The software PRAAT developed by Paul Boersma and Weenink from Amsterdam University was used to analyze the data collected. The results indicated that the identification task showed more accuracy compared with the discrimination task. The assimilation of two sounds in English (med-low and low vowels) to one sound in Portuguese (med-low vowel) confirmed the hypothesis that the absence of perceptual targets confers difficulties to the speakers in the production of sounds in a foreign language. The results confirmed the Speech Learning Model hypothesis. This research aims to contribute to the pronunciation teaching in a FL, because an accurate sound production may prevent communication problem, and it is essential for learner’s development in terms of oral communication and fluency