Examining vowel intelligibility in Brazilian Portuguese EFL learners: A study using Automatic Speech Recognition

As traditional EFL classrooms often struggle with limited time and input for pronunciation instruction, technologies such as Automatic Speech Recognition (ASR) may serve as a complementary tool to aid learners in the acquisition of specific English sounds. However, in order to consider the usefulnes...

Descripción completa

Detalles Bibliográficos
Autores: Kivisto-de Souza, Hanna, Waltelon Souza dos Santos, Fhelippe
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2024
País:Brasil
Institución:Universidade Federal de Santa Catarina (UFSC)
Repositorio:Ilha do Desterro
Idioma:inglés
OAI Identifier:oai:periodicos.ufsc.br:article/98870
Acceso en línea:https://periodicos.ufsc.br/index.php/desterro/article/view/98870
Access Level:acceso abierto
Palabra clave:L2 pronunciation
Automatic Speech Recognition
Intelligibility
Vowel production
Autonomous learning
Descripción
Sumario:As traditional EFL classrooms often struggle with limited time and input for pronunciation instruction, technologies such as Automatic Speech Recognition (ASR) may serve as a complementary tool to aid learners in the acquisition of specific English sounds. However, in order to consider the usefulness of ASR tools, more careful testing should be conducted in order to observe their ability to perform in specific contexts. The objective of the present study was to replicate Authors (2022) by examining how Microsoft Word Dictate judged the intelligibility of Brazilian Portuguese speakers of English. Contrary to Authors (2022) who examined continuous speech, the present study focused on vowels embedded in isolated words. Results showed that the intelligibility was low (50%) in comparison to Authors (2022). Discussion centers around the uses of ASR on FL pronunciation development in autonomous learning and instructed settings.