One corpus, one research question, three methods “German vowels produced by French speakers”
Abstract
A phonetic learner corpus has been created in the course of the project IFCASL (“Individualized Feedback for Computer-Assisted Language Learning”, http://www.ifcasl.org) [4, 12] which investigates the problems that arise in the French-German language pair. This corpus, for instance, allows for
the investigation of the interference French speakers have with German vowels. The data of the corpus can be analyzed with different methods (a perceptual experiment, a manual annotation and an acoustical analysis), which focus on different aspects of this problem, and which ultimately contribute different results which create a more detailed analysis of the processes in question.