Both native and nonnative language teachers often find pronunciation a difficult skill to teach because of inadequate training or uncertainty about the effectiveness of instruction. But nonnative language teachers may also see themselves as inadequate models for pronunciation, leading to increased uncertainty about whether they should teach pronunciation (Golombek & Jordan, 2005). Although studies have regularly shown that instruction is effective in promoting pronunciation improvement (Saito, 2012), it is not known if improvement depends on the native language of the instructor, nor if learners improve differently depending on whether their teacher is native or nonnative. This study investigated the effect of teachers' first language on ratings of change in accentedness and comprehensibility. Learners in intact English classes were taught one class by a nonnative-and one by a native-English-speaking teacher. Each teacher taught the same pronunciation lessons over the course of 7 weeks. Results show that native listeners' ratings of the students' comprehensibility were similar for both teachers, despite many learners' stated preference for native teachers. The results offer encouragement to nonnative teachers in teaching pronunciation, suggesting that, like other language skills, instruction on pronunciation skills is more dependent on knowledgeable teaching practices than on native pronunciation of the teacher.
In this paper, we introduce L2-ARCTIC, a speech corpus of non-native English that is intended for research in voice conversion, accent conversion, and mispronunciation detection. This initial release includes recordings from ten non-native speakers of English whose first languages (L1s) are Hindi, Korean, Mandarin, Spanish, and Arabic, each L1 containing recordings from one male and one female speaker. Each speaker recorded approximately one hour of read speech from the Carnegie Mellon University ARCTIC prompts, from which we generated orthographic and forced-aligned phonetic transcriptions. In addition, we manually annotated 150 utterances per speaker to identify three types of mispronunciation errors: substitutions, deletions, and additions, making it a valuable resource not only for research in voice conversion and accent conversion but also in computer-assisted pronunciation training. The corpus is publicly accessible at https://psi.engr.tamu.edu/l2-arctic-corpus/.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.