Comparing classifiers for pronunciation error detection

Comparing classifiers for pronunciation error detection

Strik, H.
Truong, K.
de Wet, F.
Cucchiarini, C.
TNO Defensie en Veiligheid

2007

Providing feedback on pronunciation errors in computer assisted language learning systems requires that pronunciation errors be detected automatically. In the present study we compare four types of classifiers that can be used for this purpose: two acoustic-phonetic classifiers (one of which employs linear-discriminant analysis (LDA)), a classifier based on cepstral coefficients in combination with LDA, and one based on confidence measures (the so-called Goodness Of Pronunciation scores). The best results were obtained for the two LDA classifiers which produced accuracy levels of about 85-93%. Index Terms: Computer Assisted Pronunciation Training (CAPT), pronunciation error detection, acoustic-phonetic classifiers, Goodness Of Pronunciation (GOP).

Acoustics and Audiology Informatics
Acoustic-phonetic classifiers
Computer Assisted Pronunciation Training (CAPT)
Goodness of Pronunciation (GOP)
Pronunciation error detection
Acoustics
Classifiers
Discriminant analysis
Error detection
Errors
Learning systems
Linguistics
Speech
Speech recognition
Accuracy levels
Acoustic-phonetic classifiers
Cepstral coefficients
Computer assisted language learnings
Confidence measures
Goodness of Pronunciation (GOP)
Goodness of pronunciations
LDA classifiers
Pronunciation error detection
Pronunciation error detections
Speech communication
speech

http://resolver.tudelft.nl/uuid:8da2d1e6-db00-446b-8df2-ce41adce85d2

19222

Interspeech 2007, 1837-1840

conference paper