Pronunciation Feature Extraction

Hacker, Christian; Cincarek, Tobias; Gruhn, Rainer; Steidl, Stefan; Nöth, Elmar; Niemann, Heinrich

doi:10.1007/11550518_18

Cited by 11 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a simple approach, [4] correlated mean raw segment duration values with pronunciation scores (although this metric was also highly correlated with speaking rate). [5,6] calculated the average deviations between a non-native speaker's segments and mean segment durations trained on a native speaker corpus. In another approach, [2,6] used the average log probability of segment durations based on native speaker distributions of durations for each phone (after normalizing for rate of speech).…”

Section: Introductionmentioning

confidence: 99%

Gaussian Mixture Modeling of vowel durations for automated assessment of non-native speech

Sun

Evanini

2011

2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

This paper investigates using Gaussian Mixture Model (GMM) based vowel duration features for automated assessment of non-native speech. Two different types of models were compared: a single GMM trained on a reference corpus of native speech and separate GMMs for different proficiency levels trained on a large corpus of scored non-native speech. 13 vowel categories were evaluated separately (after normalization by rate of speech), and a multiple regression model was used to evaluate the performance of all vowel categories combined. Experiments on an English language proficiency assessment show that the non-native speech GMMs outperform the native speech GMMs, and that all 13 vowels have significant correlations with human scores when the non-native speech GMMs are used. The multiple regression combination obtained correlations with human scores of 0.71 when transcriptions were used to extract the vowel durations and 0.64 when the Automatic Speech Recognition (ASR) output was used. The experiments demonstrate that the vowel duration feature based on non-native speech GMMs is a useful predictor of L2 proficiency and is robust to different datasets and situations.

show abstract

Section: Introductionmentioning

confidence: 99%

Gaussian Mixture Modeling of vowel durations for automated assessment of non-native speech

Sun

Evanini

2011

2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Using a similar method, Neumeyer et al (Neumeyer et al, 2000) designed a series of likelihood related pronunciation features, e.g., the local average likelihood and global average likelihood. Hacker et al (Hacker et al, 2005) utilized a relatively large feature vector for scoring pronunciation.…”

Section: Related Workmentioning

confidence: 99%

“…read speech or speech that is quite constrained in the use of vocabulary and syntactic structures) have emerged in the past decade (Bernstein, 1999;Witt, 1999;Hacker et al, 2005) due to the growing maturity of speech recognition and processing technologies. However, endeavors into automated scoring for spontaneous speech have been sparse given the challenge of both recognizing and assessing spontaneous speech.…”

Section: Introductionmentioning

confidence: 99%

“…Because pronunciation plays an important role in speech perception, features measuring pronuncia-tion using speech technologies have been explored in many previous studies. However, the bulk of the research on automatic pronunciation evaluation concerns read speech or highly predictable speech (Witt, 1999;Hacker et al, 2005), where there is a high possibility of success in speech recognition. Automatic pronunciation evaluation is challenging for spontaneous speech and has been under-explored.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Improved pronunciation features for construct-driven assessment of non-native spontaneous speech

Chen

Zechner

2009

Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Com

View full text Add to dashboard Cite

This paper describes research on automatic assessment of the pronunciation quality of spontaneous non-native adult speech. Since the speaking content is not known prior to the assessment, a two-stage method is developed to first recognize the speaking content based on non-native speech acoustic properties and then forced-align the recognition results with a reference acoustic model reflecting native and near-native speech properties. Features related to Hidden Markov Model likelihoods and vowel durations are extracted. Words with low recognition confidence can be excluded in the extraction of likelihood-related features to minimize erroneous alignments due to speech recognition errors. Our experiments on the TOEFL R Practice Online test, an English language assessment, suggest that the recognition/forced-alignment method can provide useful pronunciation features. Our new pronunciation features are more meaningful than an utterance-based normalized acoustic model score used in previous research from a construct point of view.

show abstract

“…Pronunciation features, as described in [10], were designed to rate a speaker's pronunciation. They are used for measuring the progress when learning a foreign language.…”

Section: Pronunciation Featuresmentioning

confidence: 99%

Analysis of Hypernasal Speech in Children with Cleft Lip and Palate

Maier

Reuß

Hacker

et al.

Text, Speech and Dialogue

Self Cite

View full text Add to dashboard Cite

Abstract. In children with cleft lip and palate speech disorders appear often. One major disorder amongst them is hypernasality. This is the first study which shows that it is possible to automatically detect hypernasality in connected speech without any invasive means. Therefore, we investigated MFCCs and pronunciation features. The pronunciation features are computed from phoneme confusion probabilities. Furthermore, we examine frame level features based on the Teager Energy operator. The classification of hypernasal speech is performed with up to 66.6 % (CL) and 86.9 % (RR) on word level. On frame level rates of 62.3 % (CL) and 90.3 % (RR) are reached.

show abstract

Pronunciation Feature Extraction

Cited by 11 publications

References 7 publications

Gaussian Mixture Modeling of vowel durations for automated assessment of non-native speech

Gaussian Mixture Modeling of vowel durations for automated assessment of non-native speech

Improved pronunciation features for construct-driven assessment of non-native spontaneous speech

Analysis of Hypernasal Speech in Children with Cleft Lip and Palate

Contact Info

Product

Resources

About