A real-time articulatory visual feedback approach with target presentation for second language pronunciation learning

Atsuo Suemitsu (1) , Jianwu Dang (1) , Takayuki Ito (2, 3) , Mark Tiede (2)

1 Japan Advanced Institute of Science and Technology
2 Haskins Laboratories
3 GIPSA-PCMD - GIPSA - Perception, Contrôle, Multimodalité et Dynamiques de la parole

Articulatory information can support learning or remediating pronunciation of a second language (L2). This paper describes an electromagnetic articulometer-based visual-feedback approach using an articulatory target presented in real-time to facilitate L2 pronunciation learning. This approach trains learners to adjust articulatory positions to match targets for a L2 vowel estimated from productions of vowels that overlap in both L1 and L2. Training of Japanese learners for the American English vowel /ae/ that included visual training improved its pronunciation regardless of whether audio training was also included. Articulatory visual feedback is shown to be an effective method for facilitating L2 pronunciation learning.

Domaines: Linguistique

Atsuo Suemitsu, Jianwu Dang, Takayuki Ito, Mark Tiede. A real-time articulatory visual feedback approach with target presentation for second language pronunciation learning. Journal of the Acoustical Society of America, 2015, 138 (4), pp.382-387. ⟨10.1121/1.4931827⟩. ⟨hal-01214661⟩