Automatic assessment of expressive oral reading

Bolaños, Daniel; Cole, Ronald A.; Ward, Wayne H.; Tindal, Gerald; Schwanenflugel, Paula J.; Kühn, Melanie

doi:10.1016/j.specom.2012.08.002

Cited by 22 publications

(19 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, their method for evaluating this required comparing children's intonation against adult intonation calculated from adult oral reading samples. Because this process can be cumbersome, we developed a similar measure of intonation contour from students' recordings based simply on the variation in pitch changes in oral readings, a measure that has proven to be a good indicator of reading expressiveness in other studies (Bolaños et al, ; Cowie et al, ). Specifically, the average pitch for the vocalic nucleus of each word was measured and a standard deviation obtained for each targeted sentence.…”

Section: Methodsmentioning

confidence: 99%

A Spectrographically Grounded Scale for Evaluating Reading Expressiveness

Benjamin

Schwanenflugel

Meisinger

et al. 2013

Reading Research Quarterly

Self Cite

View full text Add to dashboard Cite

This study evaluated the validity of a new scale for assessing children's reading fluency skill called the Comprehensive Oral Reading Fluency Scale (CORFS). The CORFS consists of two subscales that capture key elements of the Kuhn, Schwanenflugel, and Meisinger (2010) definition of reading fluency: reading expression, reading rate, and accuracy. In study 1, the reading expression subscale was developed by examining spectrographically measured prosodic differences in the oral readings of children with varying fluency skill. Three experts independently rated the oral readings of 59 second‐grade children using the CORFS. Intraclass correlations among raters were strong for both reading expression, and reading rate and accuracy. Expression ratings correlated with all but one spectrographic indicator of reading prosody and highly with standardized assessments of reading fluency and comprehension. In study 2, the structure of the scale was replicated using spectrographic measures of the oral readings of 60 third‐grade children reading two new texts. Two experts independently rated these oral readings using a final version of the scale. Results were comparable to those of study 1. Together, these results support the validity of the CORFS as an assessment of reading fluency.

show abstract

Section: Methodsmentioning

confidence: 99%

A Spectrographically Grounded Scale for Evaluating Reading Expressiveness

Benjamin

Schwanenflugel

Meisinger

et al. 2013

Reading Research Quarterly

Self Cite

View full text Add to dashboard Cite

show abstract

“…The SA evaluation includes two parts: inducing procedures and evaluation indicators. Firstly, the subject is asked to give responses following a designed inducing procedure, and then the responses are evaluated by a professional based on the selected evaluation indicators (5,7,10,19).…”

Section: Sa Evaluationmentioning

confidence: 99%

“…Specifically, SA is defined as proficiency in oral language, which includes the ability to repeat or retell properly, pronounce correctly and fluently, and express grammatical and logical content. Usually, SA is evaluated based on a collection of the subject's responses of some specific speech-inducing procedures prompted by a professional (5)(6)(7)(8)(9). To the best of our knowledge, few studies have focused on comprehensive SA evaluation for children's DLD screening, since it is incredibly challenging.…”

Section: Introductionmentioning

confidence: 99%

Fast screening for children’s developmental language disorders via comprehensive speech ability evaluation—using a novel deep learning framework

et al. 2020

View full text Add to dashboard Cite

Background: Developmental language disorders (DLDs) are the most common developmental disorders in children. For screening DLDs, speech ability (SA) is one of the most important indicators. Methods:In this paper, we propose a solution for the fast screening of children's DLDs based on a comprehensive SA evaluation and a deep framework of machine learning. Fast screening is crucial for promoting the prevalence and practicality of DLD screening which in turn is important for the treatment of DLDs and related social and behavioral abnormalities (e.g., dyslexia and autism). Our solution is focused on addressing the drawbacks existing in the previous DLD screening methods which include test failure due to text-based inducing material design and illiteracy of most young children, incomplete language evaluation indicators, and professional-reliant evaluation procedures. First, to avoid test failure, a novel comprehensive inducing procedure (CIP) with non-text (i.e., audio-visual) stimulus materials was designed that could cover a large range of modalities to adequately explore the comprehensive SA of the subjects. Second, to address incomplete language evaluation, a set of comprehensive evaluation indicators with full consideration of the characteristics of the children's language acquisition is proposed; furthermore, to break the professionalreliant limitation, we specifically designed a deep framework for fast and accurate screening.Results: Experimental results showed that the proposed deep framework is effective and professional with a 92.6% accuracy on DLD screening. Additionally, to provide a benchmark for the novel problem, we provide a CIP dataset with about 2,200 responses from over 200 children, which may also be useful for further DLD studies and insightful for the fast screening design of other behavioral abnormalities.Conclusions: Fast screening of children's DLDs can be achieved at accuracy up to 92.6% by our proposed deep learning framework. For successful fast screening, an elaborated CIP with corresponding comprehensive evaluating indicators is necessary to be designed for children suspected to have DLDs.

show abstract

“…Numerous studies have demonstrated the effectiveness of support technology on a variety of skills related to reading, including vocabulary, phonemic awareness, reading fluency, speed and comprehension [25]. Several applications have been developed not only to support the reading process but also to provide automatic assessment of oral reading by estimating errors [26], mispronunciations [27] or reading (dis)fluency [28].…”

Section: Related Workmentioning

confidence: 99%

A System for Asessing Children Readings as School

Artuso¹,

Cristoforetti

Falavigna

et al. 2017

7th ISCA Workshop on Speech and Language Technology in Education (SLaTE 2017)

View full text Add to dashboard Cite

In this paper we describe a system for analyzing the reading errors made by children of the primary and middle schools. To assess the reading skills of children in terms of reading accuracy and speed, a standard reading achievement test, developed by educational psychologists and named "Prove MT" (MT reading test), is used in the Italian schools. This test is based on a set of texts specific for different ages, from 7 to 13 years old. At present, during the test, children are asked to read aloud short stories, while teachers manually write down the reading errors on a sheet and then compute a total score based on several measures, such as duration of the whole reading, number of read syllables per second, number and type of errors, etc. The system we have developed is aimed to support the teachers in this task by automatically detecting the reading errors and estimating the needed measures. To do this we use an automatic speech-totext transcription system that employs a language model (LM) trained over the texts containing the stories to read. In addition, we embed in the LM an error model that allows to take into account typical reading errors, mostly consisting in pronunciation errors, substitutions of syllables or words, word truncation, etc. To evaluate the performance of our system we collected 20 audio recordings, uttered by 8-13 years old children, reading a novel belonging to "Prove MT" set. It is worth mentioning that the error model proposed in this paper for assessing the reading capabilities of children performs closely to an "oracle" error model obtained from manual transcriptions of the readings themselves.

show abstract

Automatic assessment of expressive oral reading

Cited by 22 publications

References 30 publications

A Spectrographically Grounded Scale for Evaluating Reading Expressiveness

A Spectrographically Grounded Scale for Evaluating Reading Expressiveness

Fast screening for children’s developmental language disorders via comprehensive speech ability evaluation—using a novel deep learning framework

A System for Asessing Children Readings as School

Contact Info

Product

Resources

About