The LeaP Corpus

This article describes the compilation, annotation and some exemplary searches of the LeaP corpus, a multilingual 12-hour corpus of spoken learner German and learner English that contains extensive time-aligned phonological and phonetic transcriptions. After outlining the collection and type of primary data that forms the corpus and explaining the manual and automatic annotation procedure, the search options of this resource are described as well as its availability for use by other researchers. The exemplary corpus search is concerned with L2 fluency. Findings of some quantitative and qualitative corpus searches are presented, in particular, the phonetic correlates of fluency and the factors that influence the fluency of foreign language learners.