Classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems
We present a system for detection of lexical stress in English words spoken by English learners. This system was designed to be part of the EduSpeak® computer-assisted language learning (CALL) software. The system uses both prosodic and spectral features to detect the level of stress (unstressed, pr...
Guardado en:
| Publicado: |
2015
|
|---|---|
| Materias: | |
| Acceso en línea: | https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_01676393_v69_n_p31_Ferrer http://hdl.handle.net/20.500.12110/paper_01676393_v69_n_p31_Ferrer |
| Aporte de: |
| id |
paper:paper_01676393_v69_n_p31_Ferrer |
|---|---|
| record_format |
dspace |
| spelling |
paper:paper_01676393_v69_n_p31_Ferrer2023-06-08T15:16:39Z Classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems Computer-assisted language learning Gaussian mixture models Lexical stress detection Mel frequency cepstral coefficients Prosodic features Computational linguistics Computer aided instruction Consumer products E-learning Feature extraction Linguistics Probability Speech recognition Computer assisted language learning Gaussian Mixture Model Mel frequency cepstral co-efficient Prosodic features Stress detection Learning systems We present a system for detection of lexical stress in English words spoken by English learners. This system was designed to be part of the EduSpeak® computer-assisted language learning (CALL) software. The system uses both prosodic and spectral features to detect the level of stress (unstressed, primary or secondary) for each syllable in a word. Features are computed on the vowels and include normalized energy, pitch, spectral tilt, and duration measurements, as well as log-posterior probabilities obtained from the frame-level mel-frequency cepstral coefficients (MFCCs). Gaussian mixture models (GMMs) are used to represent the distribution of these features for each stress class. The system is trained on utterances by L1-English children and tested on English speech from L1-English children and L1-Japanese children with variable levels of English proficiency. Since it is trained on data from L1-English speakers, the system can be used on English utterances spoken by speakers of any L1 without retraining. Furthermore, automatically determined stress patterns are used as the intended target; therefore, hand-labeling of training data is not required. This allows us to use a large amount of data for training the system. Our algorithm results in an error rate of approximately 11% on English utterances from L1-English speakers and 20% on English utterances from L1-Japanese speakers. We show that all features, both spectral and prosodic, are necessary for achievement of optimal performance on the data from L1-English speakers; MFCC log-posterior probability features are the single best set of features, followed by duration, energy, pitch and finally, spectral tilt features. For English utterances from L1-Japanese speakers, energy, MFCC log-posterior probabilities and duration are the most important features. © 2015 Elsevier B.V. 2015 https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_01676393_v69_n_p31_Ferrer http://hdl.handle.net/20.500.12110/paper_01676393_v69_n_p31_Ferrer |
| institution |
Universidad de Buenos Aires |
| institution_str |
I-28 |
| repository_str |
R-134 |
| collection |
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) |
| topic |
Computer-assisted language learning Gaussian mixture models Lexical stress detection Mel frequency cepstral coefficients Prosodic features Computational linguistics Computer aided instruction Consumer products E-learning Feature extraction Linguistics Probability Speech recognition Computer assisted language learning Gaussian Mixture Model Mel frequency cepstral co-efficient Prosodic features Stress detection Learning systems |
| spellingShingle |
Computer-assisted language learning Gaussian mixture models Lexical stress detection Mel frequency cepstral coefficients Prosodic features Computational linguistics Computer aided instruction Consumer products E-learning Feature extraction Linguistics Probability Speech recognition Computer assisted language learning Gaussian Mixture Model Mel frequency cepstral co-efficient Prosodic features Stress detection Learning systems Classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems |
| topic_facet |
Computer-assisted language learning Gaussian mixture models Lexical stress detection Mel frequency cepstral coefficients Prosodic features Computational linguistics Computer aided instruction Consumer products E-learning Feature extraction Linguistics Probability Speech recognition Computer assisted language learning Gaussian Mixture Model Mel frequency cepstral co-efficient Prosodic features Stress detection Learning systems |
| description |
We present a system for detection of lexical stress in English words spoken by English learners. This system was designed to be part of the EduSpeak® computer-assisted language learning (CALL) software. The system uses both prosodic and spectral features to detect the level of stress (unstressed, primary or secondary) for each syllable in a word. Features are computed on the vowels and include normalized energy, pitch, spectral tilt, and duration measurements, as well as log-posterior probabilities obtained from the frame-level mel-frequency cepstral coefficients (MFCCs). Gaussian mixture models (GMMs) are used to represent the distribution of these features for each stress class. The system is trained on utterances by L1-English children and tested on English speech from L1-English children and L1-Japanese children with variable levels of English proficiency. Since it is trained on data from L1-English speakers, the system can be used on English utterances spoken by speakers of any L1 without retraining. Furthermore, automatically determined stress patterns are used as the intended target; therefore, hand-labeling of training data is not required. This allows us to use a large amount of data for training the system. Our algorithm results in an error rate of approximately 11% on English utterances from L1-English speakers and 20% on English utterances from L1-Japanese speakers. We show that all features, both spectral and prosodic, are necessary for achievement of optimal performance on the data from L1-English speakers; MFCC log-posterior probability features are the single best set of features, followed by duration, energy, pitch and finally, spectral tilt features. For English utterances from L1-Japanese speakers, energy, MFCC log-posterior probabilities and duration are the most important features. © 2015 Elsevier B.V. |
| title |
Classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems |
| title_short |
Classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems |
| title_full |
Classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems |
| title_fullStr |
Classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems |
| title_full_unstemmed |
Classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems |
| title_sort |
classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems |
| publishDate |
2015 |
| url |
https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_01676393_v69_n_p31_Ferrer http://hdl.handle.net/20.500.12110/paper_01676393_v69_n_p31_Ferrer |
| _version_ |
1768542833364434944 |