Comparison of L2 Korean pronunciation error patterns from five L1 backgrounds by using automatic phonetic transcription
Abstract: This paper presents a large-scale analysis of L2 Korean pronunciation error patterns from five different language backgrounds, Chinese, Vietnamese, Japanese, Thai, and English, by using automatic phonetic transcription. For the analysis, confusion matrices are generated for each L1, by aligning canonical phone sequences and automatically transcribed phone sequences obtained from fine-tuned Wav2Vec2 XLS-R phone recognizer. Each value in the confusion matrices is compared to capture frequent common error patterns and to specify patterns unique to a certain language background. Using the Foreign Speakers' Voice Data of Korean for Artificial Intelligence Learning dataset, common error pattern types are found to be (1) substitutions of aspirated or tense consonants with plain consonants, (2) deletions of syllable-final consonants, and (3) substitutions of diphthongs with monophthongs. On the other hand, thirty-nine patterns including (1) syllable-final /l/ substitutions with /n/ for Vietnamese and (2) /\textturnm/ insertions for Japanese are discovered as language-dependent.
- P. M. Rogerson-Revell, “Computer-assisted pronunciation training (capt): Current issues and future directions,” RELC Journal, vol. 52, no. 1, pp. 189–205, 2021.
- F.-A. Chao, T.-H. Lo, T.-I. Wu, Y.-T. Sung, and B. Chen, “3m: An effective multi-view, multi-granularity, and multi-aspect modeling approach to english pronunciation assessment,” in 2022 APSIPA ASC, 2022, pp. 575–582.
- M. Yang, K. Hirschi, S. D. Looney, O. Kang, and J. H. L. Hansen, “Improving mispronunciation detection with wav2vec2-based momentum pseudo-labeling for accentedness and intelligibility assessment,” in Interspeech, 2022.
- L. Peng, K. Fu, B. Lin, D. Ke, and J. Zhang, “A study on fine-tuning wav2vec2.0 model for the task of mispronunciation detection and diagnosis,” in Interspeech, 2021.
- E. J. Kong, S. Kang, and M. Seo, “The acoustic cue-weighting and the l2 production-perception link: A case of english-speaking adults’ learning of korean stops,” Phonetics and Speech Sciences, vol. 14, no. 3, pp. 1–9, 2022.
- S. Benjamin and H.-Y. Lee, “The perception and production of korean vowels by egyptian learners,” Phonetics and Speech Sciences, vol. 13, no. 4, pp. 23–34, 2021.
- A. Jatteau, I. Vasilescu, L. Lamel, and M. Adda-Decker, “Final devoicing of fricatives in french: Studying variation in large-scale corporawith automatic alignment,” 2019.
- J. Yang and R. Fox, “L1-l2 interactions of vowel systems in young bilingual mandarin-english children,” J. Phonetics, vol. 65, pp. 60–76, 2017.
- J. Yuan, H. Lin, and Y. Liu, “Nasal coarticulation in l1 and l2 english speech: A large-scale study,” Training, pp. 96–6, 1974.
- S. Shi and C. Shih, “Acoustic analysis of l1 influence on l2 pronunciation errors: A case study of accented english speech by chinese learners,” ICPHS, Melbourne, Australia, 2019.
- S. Shi, C. Shih, and J. Zhang, “Capturing l1 influence on l2 pronunciation by simulating perceptual space using acoustic features,” in Interspeech, 2019, pp. 2648–2652.
- C. Cucchiarini and H. Strik, “Automatic phonetic transcription: An overview,” in ICPHS, 2003, pp. 347–350.
- A. Babu et al., “Xls-r: Self-supervised cross-lingual speech representation learning at scale,” in Interspeech, 2022.
- K. Choi and H.-M. Park, “Distilling a pretrained language model to a multilingual asr model,” in Interspeech, 2022.
- “Foreign speakers’ voice data of korean for artificial intelligence learning,” https://aihub.or.kr/aihubdata/data/view.do?currMenu=115&topMenu=100&aihubDataSe=realm&dataSetSn=505, 2020.
- M. McAuliffe, M. Socolof, S. Mihuc, M. Wagner, and M. Sonderegger, “Montreal forced aligner: Trainable text-speech alignment using kaldi.” in Interspeech, vol. 2017, 2017, pp. 498–502.
- D. Povey et al., “The kaldi speech recognition toolkit,” in ASRU, 2011.
- L. Wang, X. Feng, and H. Meng, “Mispronunciation detection based on cross-language phonological comparisons,” in ICALIP, 2008, pp. 307–311.
- S. Khanal, M. T. Johnson, and N. Bozorg, “Articulatory comparison of l1 and l2 speech for mispronunciation diagnosis,” in SLT, 2021, pp. 693–697.
- J. J. Holliday, “A longitudinal study of the second language acquisition of a three-way stop contrast,” Journal of Phonetics, vol. 50, pp. 1–14, 2015.
- Y. He, “Production of english syllable final/l/by mandarin chinese speakers.” Journal of Language Teaching & Research, vol. 5, no. 4, 2014.
- K. B. Choe, J. Y. Song, and S.-G. Park, “Analysis of pronunciation errors of korean syllable-final liquid /ㄹ/ by vietnamese learners of korean,” Urimal, no. 64, pp. 181–209, 2021.
- M.-k. Hwang, “Analysis of the Korean liquid pronunciation of Vietnamese entry-level Students,” Journal of Ehwa Korean Language and Literature, vol. 39, pp. 49–87, 2016.
- E. J. Lee and I. H. Woo, “A study on an educational plan for the pronunciation of the final consonants /ㄱ,ㄷ,ㅂ/ in korean for chinese korean learners －by applying the syllable structure of korean to that of chinese reversely－,” The Academy for Korean Language Education, no. 97, pp. 327–359, 2013.
- E. K. Yoon, “An experimental perceptual study of different syllable structures in l1 & l2,” Journal of Dong-ak Language and Literature, no. 64, pp. 409–438, 2015.
- L. J and P. K, “A study on the korean learners’ realization aspect and teaching methods of glottalization after obstruent,” The Korean society of bilingualism, no. 71, pp. 223–248, 2018.
- K. K H et al., “A study on the final consonant pronunciation errors of chinese korean learners,” Ph.D. dissertation, Jeju University, 2019.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.