Mova0.5 pre-release⌕
☰
home / docs / papers / English / Automatic Dialect Density Estimation for African American English

Automatic Dialect Density Estimation for African American English

What it is about

Dialect density is the share of words in an utterance that carry features of a non-standard dialect. The paper predicts it for African American English from speech. The data are about three hours of utterances from 65 speakers in five US cities, taken from CORAAL, with scores given by hand. The method combines six feature sets: ASR-transcript character bigrams and character durations, a language-model surprisal of verbs measured with a character LM trained on standard-English conversation, X-vectors and prosody embeddings projected through weak supervision, and ComParE acoustic features. An XGBoost model per feature set is then trained, and the paper reports Pearson correlation on the test split. All features together reach 0.718 on the total measure, and 0.678 averaged over 200 random splits. The most important single feature is character perplexity of the LM.

Ideas

For Mova

The chapters are 08. National varieties, 27. Negation (negative concord) and 13. Verbs (copula). The dialect plan is in The Mova dialect: a UD metadialect, treebanks, and converters. The code is en/src/tok.rs and en/src/udreg.rs. Measures: the density correlation on a held-out set with the negative control. Note: the paper's speech data (CORAAL) is not covered in the library card; check its terms before any use and treat it as evaluation only until checked.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. arXiv 2204.00967

@misc{johnson2022automatic,
  title = {{Automatic Dialect Density Estimation for African American English}},
  author = {Johnson, Alexander and Everson, Kevin and Ravi, Vijay and Gladney, Anissa and Ostendorf, Mari and Alwan, Abeer},
  year = {2022},
  eprint = {2204.00967},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2204.00967}
}

References

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

  1. J. A. Washington, L. Branum-Martin, C. Sun, and R. Lee-James, “The impact of dialect density on the growth of language and reading in African American children.” Language, speech, and hearing services in schools, 49(2), pp. 232-247. 2018. find
  2. C. Puranik, L. Branum-Martin, and J. A. Washington, J.A, “The relation between dialect density and the codevelopment of writing and reading in African American children.” Child Development, 91(4), e866–e822. PMID: 31612998 find
  3. A. Koenecke, A Nam, E. Lake, J. Nudell, M. Quartey, Z. Mengesha, C. Toups, J. R. Rickford, D. Jurafsky, and S. Goel, “Racial disparities in automated speech recognition.” Proceedings of the National Academy of Sciences Apr 2020, 117 (14) 7684-7689; DOI: 10.1073/pnas.1915768117 ↗
  4. S. Lanehart and Ayesha M. Malik. “Language use in African American communities: An introduction.” The Oxford handbook of African American language, ed. by Sonja Lanehart, pp. 1–19. Oxford: Oxford University Press. 2015. find
  5. S. Li, B. Ouyang, D. Liao, S. Xia, L. Li and Q. Hong, “End-To-End Multi-Accent Speech Recognition with Unsupervised Accent Modelling,” ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2021, pp. 6418-6422, doi: 10.1109/ICASSP39728.2021.9414833. ↗
  6. T. Tan, Y. Lu, R. Ma, S. Zhu, J. Guo and Y. Qian, “AISpeech-SJTU ASR System for the Accented English Speech Recognition Challenge,” ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2021, pp. 6413-6417, doi: 10.1109/ICASSP39728.2021.9414471. ↗
  7. R. Huang, J. H. L. Hansen and P. Angkititrakul, “Dialect/Accent Classification Using Unrestricted Audio,“ in IEEE Transactions on Audio, Speech, and Language Processing, vol. 15, no. 2, pp. 453-464, Feb. 2007, doi: 10.1109/TASL.2006.881695. ↗
  8. O. Garcı́a, and L. Wei, (2014). “Translanguaging and education. In Translanguaging: Language, bilingualism and education,” pp. 63-77. Palgrave Pivot, London, 2014. find
  9. Tyler Kendall and Charlie Farrington, “The Corpus of Regional African American Language. Version 2020.05.” Eugene, OR: The Online Resources for African American Language Project. 2020 http://oraal.uoregon.edu/coraal. ↗
  10. Alexei Baevski, Yuhao Zhou, Abdelrahman Mohamed, Michael Auli, “Wav2Vec2.0: A Framework for Self-Supervised Learning of Speech Representations.” NeurIPS 2020 find
  11. V. Panayotov, G. Chen, D. Povey, and S. Khudanpur. Librispeech: an asr corpus based on public domain audio books.” In Proc. of ICASSP, pages 5206–5210. IEEE, 2015. find
  12. A. Akbik, D. Blythe and R. Vollgraf, “Contextual String Embeddings for Sequence Labeling,” in Proceedings of the 27th International Conference on Computational Linguistics, 2018. find
  13. A. Akbik, T. Bergmann, D. Blythe, K. Rasul, S. Schweter and R. Vollgraf, “FLAIR: An easy-to-use framework for state-of-the-art NLP,” 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics (Demonstrations), 2019. find
  14. C. Cieri, D. Graff, O. Kimball, D. Miller, and K. Walker. “Fisher English Training Speech Part 1 Transcripts, LDC2004T19.” Web Download. Philadelphia: Linguistic Data Consortium, 2004. find
  15. John Hale, “A probabilistic earley parser as a psycholinguistic model”. In Proceedings of the second meeting of the North American Chapter of the Association for Computational Linguistics on Language technologies, pages 1–8. Association for Computational Linguistics, 2001. find
  16. B. Schuller, S. Steidl, A. Batliner, J. Hirschberg, J. K. Burgoon, A. Baird, A. Elkins, Y. Zhang, E. Coutinho, and K. Evanini, “The IN-TERSPEECH 2016 Computational Paralinguistics Challenge: Deception, Sincerity & Native Language,” in Proc. Interspeech, 2016, pp. 2001–2005. find
  17. F. Eyben, M. Wollmer, and B. Schuller, “Opensmile: the Munich Versatile and Fast Open-Source Audio Feature Extractor,” in Proc. ACM Multimedia, Firenze, Italy, 2010, pp. 1459–1462. find
  18. D. Snyder, D. Garcia-Romero, G. Sell, D. Povey and S. Khudanpur, “X-Vectors: Robust DNN Embeddings for Speaker Recognition,” 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018, pp. 5329-5333, doi: 10.1109/ICASSP.2018.8461375. ↗
  19. Trang Tran, Shubham Toshniwal, Mohit Bansal, Kevin Gimpel, Karen Livescu, and Mari Ostendorf. 2018. Parsing Speech: A Neural Approach to Integrating Lexical and Acoustic-Prosodic Information. In Proc. Conf. North American Chapter Assoc. for Computational Linguistics (NAACL) find
  20. Tianqi Chen and Carlos Guestrin, “XGBoost: A Scalable Tree Boosting System.” Krishnapuram, Balaji; Shah, Mohak; Smola, Alexander J.; Aggarwal, Charu C.; Shen, Dou; Rastogi, Rajeev (eds.). In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, August 13-17, 2016. ACM. pp. 785–794. arXiv:1603.02754. doi:10.1145/2939672.2939785. ↗
  21. Cynthia Puranik, Lee Branum-Martin, and Julie Washington, “The Relation Between Dialect Density and the Codevelopment of Writing and Reading in African American Children.” Child Development. 2019. 91. 10.1111/cdev.13318. ↗
  22. S. M. Lundberg and S. I. Lee, “A unified approach to interpreting model predictions,” in Proc. Adv. Neural Inf. Process. Syst., 2017, pp. 4768–4777. find
  23. H. Hermansky and N. Morgan, “RASTA processing of speech,” in IEEE Transactions on Speech and Audio Processing, vol. 2, no. 4, pp. 578-589, Oct. 1994, doi: 10.1109/89.326616. ↗

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/johnson-2022-automatic-dialect-density-estimation-african/