Automatic scoring of pronunciation quality

We present a paradigm for the automatic assessment of pronunciation quality by machine. In this scoring paradigm, both native and nonnative speech dat a is collected and a database of human-expert ratings is created to enable the development of a variety of machine scores. We first discuss issues rel ated to the design of speech databases and the reliability of human ratings . We then address pronunciation evaluation as a prediction problem, trying to predict the grade a human expert would assign to a particular skill. Usi ng the speech and the expert-ratings databases, we build statistical models and introduce different machine scores that can be used as predictor varia bles. We validate these machine scores on the Voice Interactive Language Tr aining System (VILTS) corpus, evaluating the pronunciation of American spea kers speaking French and we show that certain machine scores, like the log- posterior and the normalized duration, achieve a correlation with the targe ted human grades that is comparable to the human-to-human correlation when a sufficient amount of speech data is available. (C) 2000 Elsevier Science B.V. All rights reserved.