Speaker-based pronunciation clustering of World Englishes based on pronunciation structure analysis

摘要

English is the only language available for global communication. Due to the influence of the students' mother tongue, however, speakers from different regions inevitably have different accents in their pronunciation of English. The ultimate goal of our project is creating a global pronunciation map of World and individual Englishes, for speakers to use to locate similar English pronunciations. The speaker can then find the best English conversation partner. A learner can also know how his pronunciation geographically compares to other varieties. Creating a map mathematically requires a matrix of pronunciation distances among all the speakers considered. This paper investigates invariant pronunciation structure analysis and SVR to predict inter-speaker pronunciation distances for new speaker pairs. The speech accent archive, containing data from worldwide accented English speech, is used as training and testing samples. Experiments show very promising results.