AUTOMATIC IDENTIFICATION OF SYNTHETICALLY GENARATED INTERLANGUAGE TRANSFER PHENOMENA BETWEEN BRAZILIAN PORTUGUESE (L1) AND ENGLISH AS FOREIGN LANGUAGE (L2)

The transfer phenomena between Brazilian Portuguese and English as Foreign Language produced by non-native speakers are already known and well documented in the literature. However, the recognition and classification of these processes are made mainly through transcriptions, a slow and laborious process done by specialized linguists. The rapid identification of this type of phenomena would be of great value for software doing proficiency placement of foreign language students and could be used in language schools, distance education systems, computer-assisted pronunciation training (CAPT) or by autodidacts and linguistics researchers. The present work analyzed possible techniques and tools that can be used in the automatic simulation and identification of transfer processes. The data for the grapho-phonic-phonological transfer were synthetically generated in the Google Translate™ TTS system. Then we tested three classification algorithms to perform the identification: k-Nearest Neighbor, Centroid Minimum Distance and Artificial Neural Networks. The results indicate that these techniques are of great value to the linguistics area and for new software applications in language learning.

This repo was created to share the source code for this paper and the data used for the results.

• Authors
  • BORGES, Atos A. S.
  • RODRIGUES FILHO, Washington Luiz P.
  • ROCHA, Aratuza Rodrigues Silva
  • CARVALHO, Wilson Júnior de Araújo
  • LIMA JR., Ronaldo Mangueira
  • BARBOSA, Fábio Rocha