Elamur Research⌕
☰
home / papers / 2024 / From Rainbow to Rambo: Speaking to an AI and the Taiwanese English Pronunciation Intelligibility

From Rainbow to Rambo: Speaking to an AI and the Taiwanese English Pronunciation Intelligibility

Paper · 2024 · Aiden Yeh

Summary

The paper asks how intelligible Taiwan English (TE) pronunciation is, and it uses a machine rather than human judges as the listener. Taiwanese university students read the standard passage "Comma Gets a Cure" aloud. An AI-based automatic speech recognition (ASR) tool, Otter, turned each recording into an orthographic transcript. The author then compared these transcripts with the original text, derived intelligibility scores, and examined the phonetic sources of the misrecognised words.

Overall, TE came out as highly intelligible. The results also point to a range of sub-varieties within TE. At one end is a less prestigious basilect, and at the other is an educated acrolect close to Standard American English (SAE). This spread makes it hard to define one standard TE.

The study also has a teaching use. When students saw where the ASR misheard them, they could reflect on their own pronunciation and how comprehensible they sound. The author presents this as a way for students to treat English as part of their Taiwanese identity.

Key points

  • Speech samples from Taiwanese university students reading the "Comma Gets a Cure" passage were transcribed by the Otter ASR system and aligned with the source text, which is reproduced in an appendix.
  • Intelligibility follows Smith and Nelson's three-way split. Intelligibility means recognising words, comprehensibility means grasping meaning, and interpretability means grasping intent. The split is linked to Austin's locutionary and perlocutionary levels.
  • TE features cited from earlier work include syllable-timing and vowel insertion after consonants (stop rendered as "su-do-pu"). They also include voiced/voiceless stop mixing (p/b, t/d, k/g) and r/l confusion ("home-lun").
  • Typical ASR substitutions in the appendix:
    • vet → "bet"
    • letter → "later"
    • dog or a goat → "doc or Coke"
    • calling → "coding"
    • trap her → "Trapper"
  • The study finds a continuum from basilect to acrolect within TE, with the acrolect near SAE.

Interesting ideas

  • I1. An ASR system works as a consistent, tireless "listener" whose errors can be counted against a known reference text, which avoids rater-to-rater variation.
  • I2. A fixed, phonetically designed read-aloud passage turns open-ended speech into an aligned diagnostic, so every error is placed against a known target word.
  • I3. Error classes are defined by mechanism (voicing swap, r/l merger, inserted vowel) rather than just counted, which explains why a word failed.
  • I4. Splitting understanding into word recognition, meaning, and intent gives three separate pass/fail levels for one utterance.
  • I5. Treating a language variety as a continuum (basilect to acrolect) rather than one norm affects what counts as an "error".

For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

  • I1 (learn): Run text through a noisy channel such as TTS→ASR, then compare Mova's analyses of the clean and noisy versions. This gives a reproducible robustness test with the clean parse as the control.
  • I2 (tokens): Build a small fixed "diagnostic passage" that covers tricky tokenization, morphology, and lemma cases, and diff each new tagger build against it.
  • I3 (tokens): Normalisation for learner or ASR-derived input could include explicit, explainable rules for such substitutions (e.g. a b/v or voicing-swap candidate) instead of opaque fuzzy matching.
  • I4 (answer): Evaluation could score separately whether tokens were recovered, whether the frame was right, and whether the speech act or intent was right, showing at which layer an answer broke.
  • I5 (tree): When converting between UD dialects or parsing non-standard English, the parser could label constructions by variety instead of forcing them into standard-English analyses.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

Also on the byline: Wenzao Ursuline University of Languages.

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Arts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 worksPsychology · 1,649 worksExperimental and Cognitive Psychology · 930 works

🧭 Topics 2Research subjects and languages the work is about.English · 2,379 worksSpeech & spoken language · 519 works

Cite

Cite as Aiden Yeh. 2024. From Rainbow to Rambo: Speaking to an AI and the Taiwanese English Pronunciation Intelligibility. ELT Worldwide: Journal of English Language Teaching 11.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.26858/eltww.v11i2.57368

@article{yeh2024rainbow,
  title = {{From Rainbow to Rambo: Speaking to an AI and the Taiwanese English Pronunciation Intelligibility}},
  author = {Aiden Yeh},
  journal = {ELT Worldwide: Journal of English Language Teaching 11},
  year = {2024},
  doi = {10.26858/eltww.v11i2.57368},
  url = {https://doi.org/10.26858/eltww.v11i2.57368}
}

References

65 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Austin , J.L. (1962). How to Do Things with Words. Oxford: Oxford University Press. find
  2. Barona, D. V. (2023). Native and non-native speakers’ perceptions of non-native accents. Números, 18(2). find
  3. Chang,L. (1991). Teaching English pronunciation to speakers of Chinese (Doctoral dissertation, California State University, Northridge). find
  4. Chang, J. (2004). Ideologies of English teaching and learning in Taiwan. Department of Linguistics, Faculty of Arts, University of Sydney. find
  5. Chen, M.Y. (1975). An areal study of nasalization in Chinese. Journal of Chinese Linguistics 3.1: 16-59. find
  6. Chen, H. C. (2011). Judgments of intelligibility and foreign accent by listeners of different language backgrounds. The Journal of Asia TEFL, 8(4), 61-83. find
  7. Chiu, T. L., Liou, H. C., & Yeh, Y. (2007). A study of web-based oral activities enhanced by automatic speech recognition for EFL college learning. Computer Assisted Language Learning, 20(3), 209-233. ↗
  8. Chung, K. S. (2001). Some returned loans: Japanese loanwords in Taiwan Mandarin. Language change in East Asia, 161-179. ISSN 2303 – 3037 (Print) ISSN 2503 – 2291 (Online) find
  9. Chung, K. S. (2005). Ten core themes in pronunciation teaching. In Proceedings of the Phonetics Teaching and Learning Conference, University College London. Retrieved from http://www.phon. ucl. ac. uk/home/johnm/ptlc2005/pdf/ptlcp24. pdf. ↗
  10. Crystal, D. (2008). Two thousand million?. English today, 24(1), 3-6. ↗
  11. Dai, W. Y., & Leung, Y. N. (2005). Grades 1-9 English Teaching/Learning in Taipei, Taiwan (Exploring the Evolving Goals of English Education). JACET 全国大会要綱 , 44, 123. find
  12. Deterding, D., & Kirkpatrick, A. (2006). Emerging south‐east Asian Englishes and intelligibility. World Englishes, 25(3‐4), 391-409. ↗
  13. Gao, Z. (1995). First language phonological influences on second language acquisition (Doctoral dissertation, University of Texas at Arlington). find
  14. Golombek, P., & Jordan, S. R. (2005). Becoming “black lambs” not “parrots”: A poststructuralist orientation to intelligibility and identity. Tesol Quarterly, 39(3), 513-533. find
  15. Graham, K. M., & Yeh, Y. F. (2023). Teachers’ implementation of bilingual education in Taiwan: challenges and arrangements. Asia Pacific Education Review, 24(3), 461-472. ↗
  16. IDEA. (2019). International Dialects of English Archive. Retrieved online from https://www.dialectsarchive.com/ on March 28, 2020 ↗
  17. Hall-Lew, L. A. (2002). English loanwords in mandarin Chinese. The University of Arizona. find
  18. Ho, M. C. (1998). Culture studies and motivation in foreign and second language learning in Taiwan. Language Culture and Curriculum, 11(2), 165-182. ↗
  19. Hsieh, F. F. (2012). Low vowel raising in Sinitic languages: assimilation, reduction, or both?. Language and Linguistics, 13(4), 583. find
  20. Hsu, J. L. (2008). Glocalization and English mixing in advertising in Taiwan: Its discourse domains, linguistic patterns, cultural constraints, localized creativity, and socio-psychological effects. Journal of Creative Communications, 3(2), 155-183. find
  21. Huang, T.T. (2018). Taiwan to make English an official languages. Taiwan News. Retrieved on March 20, 2020 from https://www.taiwannews.com.tw/en/news/3515890 ↗
  22. Hubbs, E. (2013). Taiwan language-in-education policy: Social, cultural, and practical implications. The Arizona Working Papers in Second Language Acquisition and Teaching, 20, 76-95. find
  23. Jenkins, J. (2006). Current perspectives on teaching world Englishes and English as a lingua franca. TESOL quarterly, 40(1), 157-181. ↗
  24. Jian, H. L. (2004). On the syllable timing in Taiwan English. In Speech Prosody 2004, International Conference. find
  25. Jindapitak, N., Teo, A., & Savski, K. (2022). Bringing Global Englishes to the ELT classroom: English language learners’ reflections. Asian Englishes, 24(3), 279-293. ↗
  26. Kachru, B. B. (1985). Standards, codification and sociolinguistic realism: The English language in the outer circle. na. find
  27. Kachru, B. B. (1992). World Englishes: Approaches, issues and resources. Language teaching, 25(1), 1-14. ↗
  28. Kachru, B. B. (2008). The first step: The Smith paradigm for intelligibility in world Englishes. World Englishes, 27(3‐4), 293-296. ISSN 2303 – 3037 (Print) ISSN 2503 – 2291 (Online) ↗
  29. Kaiser, S. (2003). 12 Language and script in Japan and other East Asian countries: Between insularity and technology. Languages in a globalising world, 188. find
  30. Ke, I. C., & Cahyani, H. (2014). Learning to become users of English as a Lingua Franca (ELF): How ELF online communication affects Taiwanese learners' beliefs of English. System, 46, 28-38. ↗
  31. Kirkpatrick, A., Deterding, D., & Wong, J. (2008). The international intelligibility of Hong Kong English. World Englishes, 27(3‐4), 359-377. ↗
  32. Kowal, K. (2002). English spoken here. Taiwan Review, 11(02), 2002. find
  33. Kuo, I. C. (2006). Addressing the issue of teaching English as a lingua franca. ELT journal, 60(3), 213-221. ↗
  34. Kuo, J. S., Li, H., & Yang, Y. K. (2007). A phonetic similarity model for automatic extraction of transliteration pairs. ACM Transactions on Asian Language Information Processing (TALIP), 6(2), 6. ↗
  35. Li, Aijun, Qiang Fang, Ruiyuan Xu, Xuxia Wang, and Yunzhong Tang. 2005. A contrastive study between Minnan-accented Chinese and Standard Chinese. Report of Phonetic Research 18: 1-7. find
  36. Mahboob, A., & Tupas, R. (2022). Special Feature: World Englishes and Social Change. In Philippine English (pp. 363-372). Routledge. find
  37. Matsuda, A. (2003). Incorporating world Englishes in teaching English as an international language. Tesol Quarterly, 37(4), 719-729. ↗
  38. Matsuda, A. (2021). Reconceptualizing “(non-) native English Speakers” within the paradigm of teaching English as an international language. Bloomsbury World Englishes, 3, 126-142. find
  39. Morrison, W. M. (2005, November). Taiwan's accession to the WTO and its economic relations with the United States and China. Congressional Research Service, Library of Congress. find
  40. Munro, M. J., & Derwing, T. M. (1995). Foreign accent, comprehensibility, and intelligibility in the speech of second language learners. Language learning, 45(1), 73-97. ↗
  41. Nelson, C. L. (1995). Intelligibility and world Englishes in the classroom. World Englishes, 14(2), 273-279. ↗
  42. Otter. (2019). Otter Transcription Application. Retrieved online from https://otter.ai/ ↗
  43. Rajagopalan, K. (2004). The concept of ‘World English’and its implications for ELT. ELT journal, 58(2), 111-117. find
  44. Rau, D. V., Chang, H. H. A., & Tarone, E. E. (2009). Think or sink: Chinese learners' acquisition of the English voiceless interdental fricative. Language Learning, 59(3), 581-621. ↗
  45. Ren, W., Chen, Y. S., & Lin, C. Y. (2016). University students' perceptions of ELF in mainland China and Taiwan. System, 56, 13-27. ↗
  46. Rickford, Angela E., and John R. Rickford. 2007. Variation, versatility, and contrastive analysis in the classroom. In Robert Bayley and Ceil Lucas (eds), Sociolinguistic variation: Theories, methods and applications, 276-296. Cambridge: Cambridge University Press. ↗
  47. Smith, L. E., & Nelson, C. L. (1985). International intelligibility of English: Directions and resources. World Englishes, 4(3), 333-342. ISSN 2303 – 3037 (Print) ISSN 2503 – 2291 (Online) ↗
  48. Smith, L. E. (1992). Spread of English and issues of intelligibility. The other tongue: English across cultures, 2, 75-90. find
  49. Smith, Larry E., and Nelson, Cecil L. (2006) World Englishes and issues of intelligibility. In Braj B. Kachru, Yamuna Kachru, and Cecil L. Nelson (eds.), The Handbook of World Englishes (pp. 428–5). Malden, MA: Blackwell. find
  50. Tupas, R. (2020). Decentering language: Displacing Englishes from the study of Englishes. Critical Inquiry in Language Studies, 17(3), 228-245. ↗
  51. Upton, Clive. 2008. Received pronunciation. In B. Kortmann and C. Upton (eds). Varieties of English 1: The British isles, 237-252. Berlin: Mouton de Gruyter. find
  52. Visceglia, T., Tseng, C. Y., Su, Z. Y., & Huang, C. F. (2010). Interaction of lexical and sentence prosody in Taiwan L2 English. In Second Language Studies: Acquisition, Learning, Education and Technology. find
  53. Wang, L. Y. (2021). English language education in the national development planning of modern Taiwan. International Journal of Taiwan Studies, 5(1), 42-58. ↗
  54. Wassink, A. B. (1999). Historic low prestige and seeds of change: Attitudes toward Jamaican Creole. Language in Society, 28(1), 57-92. ↗
  55. Widdowson, H. G. (1994). The ownership of English. TESOL Quarterly, 28(2), 377‐389. ↗
  56. Wilkinson, J. (2020). From Native Speaker to Intercultural Speaker and Beyond. The Routledge Handbook of Language and Intercultural Communication, 283-298. ↗
  57. Williams, C. (2007). Research methods. Journal of Business & Economics Research (JBER), 5(3). find
  58. Wolfram, W., & Schilling-Estes, N. (2004). Remnant dialects in the coastal United States. Legacies of colonial English: Studies in transported dialects, 172-202. ↗
  59. Wu, J. R. (2012). GEPT and English language teaching and testing in Taiwan. Language Assessment Quarterly, 9(1), 11-25. ↗
  60. Wu, K. H., & Ke, C. (2009). Haunting Native Speakerism? Students' Perceptions toward Native Speaking English Teachers in Taiwan. English Language Teaching, 2(3), 44-52. find
  61. Yang, J. H. (2012). A socio-phonological analysis of Taiwan English from the perspective of world Englishes. Taiwan Journal of Linguistics, 10(1), 115-142. find
  62. Yeh, A. (2023). Bilingual 2030: Revving Up English in a Multilingual, Multicultural Taiwan. Asian Englishes, 1-15. ↗
  63. Zee, Eric. 1985. Sound change in syllable-final nasal consonants in Chinese. Journal of Chinese Linguistics 13: 291-330. find
  64. Zhang, L. (2015). An empirical study on the intelligibility of English Spoken by Chinese university students. Chinese Journal of Applied Linguistics, 38(1), 36-54. ↗
  65. Zhou, W. J. & Deterding, D. & Nolan, F. (2019). Intelligibility in Chinese English Spoken in Central China. Chinese Journal of Applied Linguistics. 42. 449-465. 10.1515/CJAL-2019-0027. ↗

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/yeh-2024-rainbow-rambo-speaking-ai-taiwanese/