본문 바로가기
KERD
논문KCI 등재

따라읽기 교육 시스템 개발용 끊어읽기 경계 주석 코퍼스 구축에 관한 예비 연구

A Pilot Study on Building a Chunk Boundary -annotated English Corpus for a Computer-assisted Echo Reading Education System

김선희|임서현
사단법인 한국언어학회 (The Linguistic Society of Korea)|2022.04|언어학 (EONEOHAG)|제92호|pp. 3-40|DOI 10.17290/jlsk.2022..92.3

0

조회

0

다운로드

0

인용

0

참고문헌

키워드

chunkingpausephrasingecho readinginter-transcriber agreement
주제분류언어학

초록

This paper presents a pilot study on constructing a chunk boundary-annotated English corpus to develop a computer-assisted echo reading education system for beginner English learners. Echo reading is an effective reading strategy in which the teacher reads a chunk of text aloud and students echo back the reading of the chunk. To construct the corpus, it is assumed that the research participants, adult English speakers, teach English to beginner learners of English using the echo reading strategy. In the corpus, a chunk has 2~4 words (3~5 syllables) on average and sentence length and pause number are highly correlated. When they teach echo reading to beginner learners of English, the adult speakers of English can choose their strategies from three options: changing the number of chunks, the chunk length, or both, if the length of a sentence changes. The participants of the experiment showed individual differences in chunking for echo reading education by selecting different strategies. We need to consider the most appropriate chunking strategies in building a chunk boundary-annotated corpus for echo reading education for beginner English learners. 69.05% of the chunking boundaries in the small corpus are identical to the syntactic phrase boundaries. Only 59.88% of the semantic chunk boundaries are identical to the chunking boundaries. In addition, English speakers pause after a comma, but they do not pause when it is used for word or phrase listing. They also put a chunk boundary after a clause. On the other hand, there tend to be no chunk boundaries after function words. If a subject’s syllables are less than 4 syllables, English speakers do not pause after the subject. The chunk boundary-annotated corpus we develop should reflect these chunking patterns. The chunk-annotated corpus for a computer-assisted echo-reading system will be useful in various research fields such as English education, linguistics, Spoken Language Processing, and Natural Language Processing.

Abstract

This paper presents a pilot study on constructing a chunk boundary-annotated English corpus to develop a computer-assisted echo reading education system for beginner English learners. Echo reading is an effective reading strategy in which the teacher reads a chunk of text aloud and students echo back the reading of the chunk. To construct the corpus, it is assumed that the research participants, adult English speakers, teach English to beginner learners of English using the echo reading strategy. In the corpus, a chunk has 2~4 words (3~5 syllables) on average and sentence length and pause number are highly correlated. When they teach echo reading to beginner learners of English, the adult speakers of English can choose their strategies from three options: changing the number of chunks, the chunk length, or both, if the length of a sentence changes. The participants of the experiment showed individual differences in chunking for echo reading education by selecting different strategies. We need to consider the most appropriate chunking strategies in building a chunk boundary-annotated corpus for echo reading education for beginner English learners. 69.05% of the chunking boundaries in the small corpus are identical to the syntactic phrase boundaries. Only 59.88% of the semantic chunk boundaries are identical to the chunking boundaries. In addition, English speakers pause after a comma, but they do not pause when it is used for word or phrase listing. They also put a chunk boundary after a clause. On the other hand, there tend to be no chunk boundaries after function words. If a subject’s syllables are less than 4 syllables, English speakers do not pause after the subject. The chunk boundary-annotated corpus we develop should reflect these chunking patterns. The chunk-annotated corpus for a computer-assisted echo-reading system will be useful in various research fields such as English education, linguistics, Spoken Language Processing, and Natural Language Processing.

참고문헌 (64)

  1. 김지연, "문법 단위로서 사고 단위에 기반을 둔 초등 예비교사의 영어 문장 끊어 읽기와 발음 능력 향상", 영어영문학, 26(3), pp.201~232 (2021), 미래영어영문학회[DOI]
  2. 김현지, "학령전기 아동과 성인의 과업에 따른 구절나누기 특성", 언어치료연구, 28(1), pp.77~86 (2019), 한국언어치료학회[DOI]
  3. 김희경, "한국인의 영어 운율구 구현 양상- 경계성조를 중심으로 -", 새한영어영문학, 51(4), pp.185~208 (2009), 새한영어영문학회[DOI]
  4. 박신애, "중학생의 읽기 유창성 양상 연구", 리터러시 연구, 10(3), pp.487~519 (2019), 한국 리터러시 학회[DOI]
  5. 신혜연, "정상 성인의 구절나누기 특징", 한국언어치료학회 학술발표대회 논문집, 42, pp.426~428 (2016)
  6. 안종복, "정상적인 쉼(pause)의 길이와 빈도에 대한 규준 확립을 위한 예비 연구", 언어치료연구, 14(3), pp.229~237 (2005), 한국언어치료학회
  7. 윤미선, "청각장애인과 건청인의 말 속도와 쉼 특성 비교", Communication Sciences and Disorders, 9(1), pp.15~29 (2004), 한국언어청각임상학회
  8. 이윤, "초등학생 영어 읽기에서 운율감과 독해력의 상관 분석", Foreign Languages Education, 22(1), pp.75~97 (2015), 한국외국어교육학회
  9. 홍선영, "낭독 훈련이 한국인 영어 학습자 발화의 휴지 실현 양상에 미치는 영향", 인문논총, 50, pp.103~125 (2019), 인문과학연구소[DOI]
  10. 홍선영, "한국인 영어 학습자의 운율구 발화와 학습에 관한 연구", 영미어문학, (135), pp.187~204 (2019), 한국영미어문학회
  11. Bai, Y., "ASR-based Evaluation and Feedback for Individualized Reading Practice", INTERSPEECH 2020 (2020)
  12. Bai, Y., "An ASR-based Reading Tutor for Practicing Reading Skills in the First Grade : Improving Performance through Threshold Adjustment", IberSPEECH (2021)
  13. Bishop, J., "Individual Difference in Prosodic Strategies to Sentence Parsing", ICPhS, 2015 (2015)
  14. Breen, M., "Inter-Transcriber Reliability for Two Systems of Prosodic Annotation : ToBI(Tones and Break Indices)and RaP(Rhythm and Pitch)", Corpus Linguistics and Linguistic Theory, 8(2), pp.277~312 (2012)
  15. Canavan, A., "CALLHOME American English Speech LDC97S42", Linguistic Data Consortium (1997)
  16. Chen, S., "On Rhythem, Prosodic Grouping, and Declination Pattern of Taiwan Mandarin Learners of English", the 10th International Conference on Speech Prosody 2020 (2020)
  17. Demenko, G., "An Audiovisual Feedback System for Acquiring L2 Pronunciation and L2 Prosody", Proceedings of the International Workshop on Speech and Language Technology in Education 2009 (2009)
  18. Durski, S., "Reading Fluency Training with Amazon Alexa", Proceedings of the 12th International Conference on Computer Supported Education (CSEDU 2020), 2, pp.139~146 (2020)
  19. Escudero, D., "Analysis of Inter-transcriber Consistency in the Cat_ToBI Prosodic Labeling System", Speech Communication, 54(5), pp.566~582 (2012)
  20. Fodor, J. D., "Prosodic Disambiguation in Silent Reading", North East Linguistic Society, 32(8) (2002)
  21. Frank, M., "Reading Fluency in Children and Adolescents Who Stutter", Brain Sciences, 11(1), pp.1595~ (2021)
  22. Frazier, L., "Prosodic Phrasing is Central to Language Comprehension", TRENDS in Cognitive Science, 10(6) (2006)
  23. Frota, S., "(Dys)Prosody in Parkinsons’ Disease : Effects of Medication and Disease Duration on Intonation and Prosodic Phrasing", Brain Sciences, 11(8), pp.1100~ (2021)
  24. Godde, E., "Reading Prosody Development: Automatic Assessment for a Longitudinal Study", SLaTE2019 – The 8th ISCA Workshop on Speech and Language Technology in Education (2019)
  25. Erika Godde, "Pausing and breathing while reading aloud: development from 2nd to 7th grade in French speaking children", Reading and Writing, 35(1), pp.1~27 (2021), Springer Science and Business Media LLC[DOI]
  26. Godfrey, J. J., "Switchboard-1 Release 2 LDC97S62", Linguistic Data Consortium (1993)
  27. Lindsay Heggie, "Prosodic awareness and punctuation ability in adult readers", Reading Psychology, 39(2), pp.188~215 (2018), Informa UK Limited[DOI]
  28. Huang, B. H., "The Effect of Age on the Acquisition of Second Language Prosody", Language and Speech, 54(3), pp.387~414 (2011)
  29. Hussein, Z. M., "The Impact of Using Chunking Technique on Developing Reading Skill and Perception of Intermedicate School Students", Indian Journal of Public Health Research and Development, 10(6) (2019)
  30. Joshi, V., "Extending a Parser to Distant Domains Using a Few Dozen Partially Annotated Examples", ACL 2018 (2018)
  31. Jouaiti, M., "Dysfluency Classification in Stuttered Speech using Deep Learning for Real-Time Applications", ICASSP 2022 (2022)
  32. Kim, J., "Individual Differences in the Production of Prosodic Boundaries in American English", International Congress of Phonetic Sciences 2019 (2019)
  33. Minho Kim, "Korean Prosody Phrase Boundary Prediction Model for Speech Synthesis Service in Smart Healthcare", Electronics, 10(19), pp.2371~ (2021), MDPI AG[DOI]
  34. Landis, J. R., "The Measurement of Observer Agreement for Categorical Data", Biometrics, 1977, pp.159~174 (1977)
  35. Li, M., "An Automated Assessment Framework for Atypical Prosody and Stereotyped Idiosyncratic Phrases Related to Autism Spectrum Disorder", Computer Speech & Language, 56(2019), pp.80~94 (2017)
  36. Limonard, S., "Analyzing Read Aloud Speech by Primary School Pupils: Insights for Research and Development", INTERSPEECH 2020 (2020)
  37. Theresa Matzinger, "Non-native speaker pause patterns closely correspond to those of native speakers at different speech rates", PLOS ONE, 15(4), pp.e0230710~ (2020), Public Library of Science (PLoS)[DOI]
  38. McCauley, S. M., "Chunking Ability Shapes Sentence Processing at Multiple Levels of Abstraction", CogSci 2017 (2017)
  39. Nishida, H., "The Influence of Chunking on Reading Comprehension : Investigating the Acquisition of Chunking Skill", The Journal of ASIA TEFL, 10(4), pp.163~183 (2013)
  40. Ostendorf M., "The Boston University Radio News Corpus", Linguistic Data Consortium, pp.1~19 (1995)
  41. Randolph, J. J., "Free-marginal Multirater Kappa: An Alternative to Fleiss’Fixed-marginal Multirater Kappa", Paper Presented at the Joensuu University Learning and Instruction Symposium 2005 (2005)
  42. (2008)[URL]
  43. Rocholl, J. C., "Disfluency Detection with Unlabeled Data and Small BERT Models", arXiv preprint arXiv:2104.1076, ~ (2021)
  44. Romana, A., "Automatically Detecting Errors and Disfluencies in Read Speech to Predict Cognitive Impairment in People with Parkinson’s Disease", INTERSPEECH 2021 (2021)
  45. Rosel, A. S., "The Influence of Echo Reading Strategy towards Students’Reading Comprehension in Descriptive Text at the First Semester of the Eighth Grade of SMP Tamansiswa Teluk Betung in the Academic Year of 2017", 2018, UIN Raden Intan Lampung, 2018
  46. Rosenberg, A., "Automatic Detection and Classification of Prosodic Events", PhD, pp.2009
  47. Rosenberg, A., "Modeling Phrasing and Prominence Using Deep Recurrent Learning", Proceedings of the Sixteenth Annual Conference of the international Speech Communication Association, 2015 (2015)
  48. Joseph Roy, "Individual differences and patterns of convergence in prosody perception", Laboratory Phonology: Journal of the Association for Laboratory Phonology, 8(1), pp.22~ (2017), Open Library of the Humanities[DOI]
  49. Sabu, K., "Prosodic Event Detection in Children’s Read Speech", Computer Speech & Language, 2021 (2021)
  50. Sankaranarayanan, A., "Automatic Prosodic Event Detection Using Acoustic, Lexical, and Syntactic Evidence", IEEE Trans Audio Speech Lang Processing, 16(1), pp.216~228 (2008)
  51. Septiana, R., "Teaching Reading Narrative Texts through Chunking Strategy", English Empower, 3(2) (2019)
  52. Shi, P., "Simple BERT Models for Relation Extraction and Semantic Role Labeling", arXiv preprint arXiv:1904.05255, ~ (2019)
  53. Shinozaki, F., "What Sentence Structures are Improved by Chunk-based Exercises?", Language, Individual & Society, Journal of International Scientific Publications, 11 (2017)
  54. Sloan, R., "Prosody Prediction from Syntactic, Lexical, and Word Embedding Features", the 10th ISCA Workshop on Speech Synthesis (SSW 10) (2019)
  55. Steinhauer, K., "Electrophysiological Correlates of Prosody and Punctuation", Brain and Language, 86(1), pp.142~164 (2003)
  56. Steinhauer, K., "Prosodic Boundaries, Comma Rules, and Brain Responses : The Closure Positive Shift in ERPs as a Universal Marker for Prosodic Phrasing in Listeners and Readers", Journal of Psycholinguistic Research, 30(3), pp.2001~ (2001)
  57. Suzuki, S., "The Effects of Read-aloud Assistance on Second Language Oral Fluency in Text Summary Speech", Proceedings of DiSS 2019, pp.31~34 (2019)
  58. Taylor, P., "Assigning Phrase Breaks from Part-Of-Speech Sequences", Computer Speech and Language, 12, pp.99~117 (1998)
  59. Tran, T., "Analysis of Disfluency in Children’s Speech", arXiv preprint arXiv:2010.04293, ~ (2020)
  60. 윤태진, "Speaker variation in English prosodic boundary", 언어연구, 31(1), pp.1~23 (2014), 언어정보연구소[DOI]
  61. Yoon, T. -J., "Intertranscriber Reliability of Prosodic Labeling on Telephone Conversation using ToBI", Proceedings of the 8th International Conference on Spoken Language Processing (2004)
  62. Zhao, G., "L2-ARCTIC : A Non-Native English Speech Corpus", INTERSPEECH 2018 (2018)
  63. Prosody in English for Development of a Reading Tutor, "Zhao, X.", ~, pp.IPSJ SIG Technical Report Vol. 2018-SLP-125, No. 17 (2018)
  64. Zhou, T., "An Automated Assessment Framework for Speech Abnormalies Related to Autism Spectrum Disorder", The 3rd International Workshop on Affective Social Multimedia Computing (ASMMC), 2017 (2017)

관련 논문AI추천

영어 화자의 be-동사 앞/뒤 끊어 읽기 패턴에 관한 인지적 직관 연구: 컴퓨터 지원 영어 읽기 교육을 위한 기초 실험 연구김선희, 임서현한국언어학회 · 언어2021 · 제46권 제2호문법 단위로서 사고 단위에 기반을 둔 초등 예비교사의 영어 문장 끊어 읽기와 발음 능력 향상김지연한국영어영문학회 · 영어영문학2021 · 제26권 제3호한국어 학습자의 소리 내어 읽기와 문해력 간의 상관관계 연구-의미 단위 띄어 읽기 및 오독 분석을 중심으로-윤나영한국언어문화교육학회 · 언어와 문화2022 · 제18권 제3호The Difference in Pausing Intervals between Korean Learners of English and Native Speakers of English이용철사단법인 한국언어학회 · 언어학2007 · 제15권 제3호초등학교 2학년 학습자의 띄어 읽기 양상 연구이관희, 최선희초등교육연구원 · 한국초등교육2025 · 제36권 제3호한국 중학생의 영어 읽기 발화에서 문장유형에 따른 유창성 등급과 초분절 요소의 관계김화영한국음성학회 · 말소리와 음성과학2022 · 제14권 제3호

이 간행물의 다른 논문 (1건)

한국어 음운 연구의 경향과 전망— 국어학과 언어학 분야의 비교를 중심으로 —박선우pp. 41-64
원문 보기