Skip to main content

Automatic assessment of non-native accent degrees using phonetic level posterior and duration features from multiple languages

Publication ,  Conference
Chen, S; Zhou, Y; Li, M
Published in: 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015
February 19, 2016

This paper presents an automatic non-native accent assessment approach using phonetic level posterior and duration features. In this method, instead of using conventional MFCC trained Gaussian Mixture Models (GMM), we use phonetic phoneme states as tokens to calculate the posterior probability and zero-oder Baum-Welch statistics. Phoneme recognizers from five languages are employed to extract phonetic level features. It is shown that features based on these five languages' phoneme recognizers are complementary for capturing non-native information and phoneme duration based features are most effective in this task. The final proposed fusion system achieved 0.6089 Spearman's Correlation Coefficient on the test set, which outperformed the openSMILE baseline by 43.3%.

Duke Scholars

Published In

2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015

DOI

Publication Date

February 19, 2016

Start / End Page

156 / 159
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Chen, S., Zhou, Y., & Li, M. (2016). Automatic assessment of non-native accent degrees using phonetic level posterior and duration features from multiple languages. In 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015 (pp. 156–159). https://doi.org/10.1109/APSIPA.2015.7415493
Chen, S., Y. Zhou, and M. Li. “Automatic assessment of non-native accent degrees using phonetic level posterior and duration features from multiple languages.” In 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015, 156–59, 2016. https://doi.org/10.1109/APSIPA.2015.7415493.
Chen S, Zhou Y, Li M. Automatic assessment of non-native accent degrees using phonetic level posterior and duration features from multiple languages. In: 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015. 2016. p. 156–9.
Chen, S., et al. “Automatic assessment of non-native accent degrees using phonetic level posterior and duration features from multiple languages.” 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015, 2016, pp. 156–59. Scopus, doi:10.1109/APSIPA.2015.7415493.
Chen S, Zhou Y, Li M. Automatic assessment of non-native accent degrees using phonetic level posterior and duration features from multiple languages. 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015. 2016. p. 156–159.

Published In

2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015

DOI

Publication Date

February 19, 2016

Start / End Page

156 / 159