A method for measuring the intelligibility and nonnativeness of phone quality in foreign language pronunciation training

Kawai, Goh; Hirose, Keikichi

doi:10.21437/icslp.1998-752

Cited by 13 publications

(5 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this sense our situation is different from that o f many studies on the use o f ASR in automatic pronunciation assessment, in which fixed language pairs (L1 & L2 fixed) are involved [e.g. 1,6]. In our case L2 is always Dutch, but the L1 o f the language students are extremely diverse.…”

Section: Introductionmentioning

confidence: 75%

“…First, it is difficult to determine exactly what an anti-model should represent if the target utterance is known to be produced by someone learning Dutch as a second language. Other than in most other studies reported on in this field [1,6], there is an enormous diversity in the language backgrounds o f the subjects whose Dutch oral proficiency needs to be evaluated by our system. Secondly, even if it were possible to clearly define such an anti-model, the availability o f a sufficient amount o f applicable training material would still remain an unresolved issue.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Using likelihood ratios to perform utterance verification in automatic pronunciation assessment

Wet¹,

Cucchiarini²,

Strik³

et al. 1999

6th European Conference on Speech Communication and Technology

View full text Add to dashboard Cite

The aim of our current research is to investigate the possibility of using likelihood ratios to perform utterance verification within the context of automatic oral proficiency assessment. The likelihood ratios under investigation have the appealing feature that they may be computed simply by using an off-theshelf automatic speech recognition system in two different recognition modes (forced and free phone) instead of using a system with specifically trained anti-models. We achieved 93% correct classification for 10 phonetically rich sentences uttered by 60 non-native language students.

show abstract

Section: Introductionmentioning

confidence: 75%

Section: Methodsmentioning

confidence: 99%

Using likelihood ratios to perform utterance verification in automatic pronunciation assessment

Wet¹,

Cucchiarini²,

Strik³

et al. 1999

6th European Conference on Speech Communication and Technology

View full text Add to dashboard Cite

show abstract

“…The problems with this method, however, are that HMMs are not good at distinguishing between phonemes that differ mainly in temporal structure and that it does not take into account effects from speaking rate. In Kawai et al [4], they employed listening tests for developing a method. In these tests, they used minimal pairs differentiated by the length of one vowel such as /to:ru/ (to pass) and /toru/ (to take).…”

Section: Research On Automatic Classification Of Vowel Lengthmentioning

confidence: 99%

“…For such an approach, better understanding of the perception of vowel length is needed. For this, listening tests like the ones in Kawai et al [4] and Yamamoto et al's [5] works can be used to further investigate the mechanism of perception.…”

Section: Overviewmentioning

confidence: 99%

Automatic recognition of vowel length in Japanese for a CALL system motivated by perceptual experiments

Short,

Hirose,

Minematsu

2013

Speech and Language Technology in Education (SLaTE 2013)

View full text Add to dashboard Cite

Acquisition of the Japanese vowel length contrast can be problematic for non-native speakers. For these speakers, a CALL system which can automatically recognize vowel length could be of great benefit for pointing out their errors and issuing corrective feedback. However, a method that can adequately do this has not been proposed yet. Vowel length recognition is made difficult because the vowel length distinction is dependent on the surrounding vowel durations which vary due to speaking rate and other factors. Hidden Markov Models (HMMs), the standard way of recognizing this distinction, do not make use of this information. Methods have been proposed to recognize this in the past, but they do not appear viable unless knowledge about the durations of other vowels is present. Thus, we carry out perceptual experiments to gain more knowledge about the vowel length contrast. From this analysis, we develop an automatic recognition algorithm for vowel length that uses support vector machines (SVMs). We tested this method on a speaking rate corpus, native speech, and non-native speech. The method produced recognition results that are overall superior to HMMs and also more robust against speaking rate differences with an average of a 0.83 correct recognition rate for the 3 datasets. The error and non-error classification rates on non-native speech for this are 0.86 and 0.84 respectively.

show abstract

“…Regarding the latter, two different approaches can be adopted: "model merging" and "parallel models." In the "parallel models" approach, acoustic models for both languages are stored, and during decoding the recognizer determines which models fit the data better [24][25][26][27]. In the "model merging" (or model interpolation) approach, acous tic models ofboth languages are combined, in order to obtain a new set of acoustic models [26].…”

Section: Non-native Speechmentioning

confidence: 99%

Optimizing Automatic Speech Recognition for Low-Proficient Non-Native Speakers

Doremalen

Cucchiarini

Strik

2010

EURASIP Journal on Audio, Speech, and Music Processing

View full text Add to dashboard Cite

Computer-Assisted Language Learning (CALL) applications for improving the oral skills of low-proficient learners have to cope with non-native speech that is particularly challenging. Since unconstrained non-native ASR is still problematic, a possible solution is to elicit constrained responses from the learners. In this paper, we describe experiments aimed at selecting utterances from lists of responses. The first experiment on utterance selection indicates that the decoding process can be improved by optimizing the language model and the acoustic models, thus reducing the utterance error rate from 29-26% to 10-8%. Since giving feedback on incorrectly recognized utterances is confusing, we verify the correctness of the utterance before providing feedback. The results of the second experiment on utterance verification indicate that combining duration-related features with a likelihood ratio (LR) yield an equal error rate (EER) of 10.3%, which is significantly better than the EER for the other measures in isolation. 2EURASIP Journal on Audio, Speech, and Music Processing distinction between phonemes in the target language, thus producing one phoneme instead of two distinct ones. This is the case with many non-native speakers of English, for instance, Germans [6], who have difficulty in realizing the distinction between the English phonemes /ae/ and /e/ and often produce /e/ when /ae/ should be used, or Japanese speakers of English who have difficulty in distinguishing /l/ and /r/ [7] and may end up producing sounds that are neither an English /l/ nor an English /r/. In such cases, confusion may arise because distinct words will be realized in the same way. This can also happen when speech sounds are inappropriately deleted or inserted, which is another common phenomenon in non-native speech [8].With respect to morphology and syntax the speech of non-natives may also exhibit deviations from that of native speakers [9]. At the level of morphology, they may find it difficult to produce correct forms of verbs, nouns, adjectives, articles, and so forth, especially when the morphological distinction hinges on subtle phonetic distinctions, such as the presence of a plosive or fricative sound in consonant clusters or the distinction between two similar vowels (lead versus led). Irregular verbs and nouns may also pose serious problems, resulting in the production of nonexistent regularized forms. Deviations in syntax may concern the structure of sentences, the ordering of constituents and their omission or insertion. As to vocabulary, non-native speakers also tend to have a limited and often deviant lexicon. Finally, non-native speech exhibits more disfluencies and hesitation phenomena than native speech and is characterized by a lower speech rate [10][11][12][13][14].All these problems are compounded when dealing with speech of non-natives that are still in the process of learning the language. In general, the degree of deviation from native speech and the incidence of disfluencies will be in inverse relation to the degree...

show abstract

A method for measuring the intelligibility and nonnativeness of phone quality in foreign language pronunciation training

Cited by 13 publications

References 2 publications

Using likelihood ratios to perform utterance verification in automatic pronunciation assessment

Using likelihood ratios to perform utterance verification in automatic pronunciation assessment

Automatic recognition of vowel length in Japanese for a CALL system motivated by perceptual experiments

Optimizing Automatic Speech Recognition for Low-Proficient Non-Native Speakers

Contact Info

Product

Resources

About