In Computer Assisted Language Learning systems, pronunciation scoring consists in providing a score grading the overall pronunciation quality of the speech uttered by a student. In this work, a log-likelihood ratio obtained with respect to two automatic speech recognition (ASR) models was used as score. One model represents native pronunciation while the other one captures non-native pronunciation. Different approaches to obtain each model and different amounts of training data were analyzed. The best results were obtained training an ASR system using a separate large corpus without pronunciation quality annotations and then adapting it to the native and non-native data, sequentially. Nevertheless, when models are trained directly on the native and non-native data, pronunciation scoring performance is similar. This is a surprising result considering that word error rates for these models are significantly worse, indicating that ASR performance is not a good predictor of pronunciation scoring performance on this system.
CITATION STYLE
Landini, F., Ferrer, L., & Franco, H. (2017). Adaptation approaches for pronunciation scoring with sparse training data. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10458 LNAI, pp. 87–97). Springer Verlag. https://doi.org/10.1007/978-3-319-66429-3_8
Mendeley helps you to discover research relevant for your work.