The Maaloula Aramaic Speech Corpus (MASC): From Printed Material to a Lemmatized and Time-Aligned Corpus

What it is about

The corpus turns the printed transcriptions of 173 tape-recorded narratives in Maaloula Aramaic, a Western Neo-Aramaic variety of Syria, into a usable resource. It contains 64,845 word tokens from 45 speakers. The printed texts were scanned with OCR (which does not know the language, so it drops diacritics), corrected by a native consultant against the audio, and marked for the author's errors with [sic] and for loanwords with <ar> tags. A hand-made list of 12,220 word forms was mapped to 3,781 lemmas from 1,932 roots using the author's dictionary. The audio (about 10 hours) was denoised and aligned to the text with WebMAUS, giving four tiers in Praat. The corpus is offered in four formats, including an SQLite database with eight linked tables. The authors also show one study that used the corpus to test a generalisation on vowel epenthesis.

Ideas

• I1. A table of deterministic OCR corrections for an Aramaic printed edition: the substitutions S to ʕ, c to č, ö to ō and ô to ō. The paper shows that other confusions (š and ṣ, ḳ and k, ḥ and h, where the OCR drops the diacritic) cannot be fixed automatically, so the table should be paired with a list of words for hand correction. This goes into tools/ocrfix.
• I2. Mark interrupted words, false starts and speaker errors with separate tags that the tokenizer reads as classes, not as lemmas. The paper keeps "--" for interrupted words and "..." for false starts, and this distinction matters for the parser's input.
• I3. A root search as a test for the lexicon: the wildcard search for the root of "to become" in the lemma list finds 197 forms in the text, of which some are false hits from a different root. A rule with the root fixed by the lemma table gives a gold test of root extraction in The lexicon of Biblical Aramaic.

For Mova

I1 goes into tools/ocrfix as an Aramaic OCR case. I2 is a tagging rule for tools/langseg and the Aramaic tokenizer. I3 is a test for The lexicon of Biblical Aramaic and for the root table of The Aramaic family and the Jewish varieties. The corpus itself is licensed NC and is for tests only, so no text from it is carried into a release.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.63317/2sbg63r7bc8y

@inproceedings{eid2022maaloula,
  title = {{The Maaloula Aramaic Speech Corpus (MASC): From Printed Material to a Lemmatized and Time-Aligned Corpus}},
  author = {Eid, Ghattas and Seyffarth, Esther and Plag, Ingo},
  booktitle = {Proceedings of the Thirteenth Language Resources and Evaluation Conference},
  year = {2022},
  doi = {10.63317/2sbg63r7bc8y},
  url = {https://doi.org/10.63317/2sbg63r7bc8y}
}

References

26 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Anthony, L. (2020). AntConc [Computer program] (Version 3.5.9) [Computer software]. Version 3.5.9, retrieved from https://www.laurenceanthony.net/software ↗
2. Arnold, W. (1991a). Das Neuwestaramäische: III. Volkskundliche Texte aus Maʿlūla. Otto Harrassowitz. find
3. Arnold, W. (1991b). Das Neuwestaramäische: IV. Orale Literatur aus Maʿlūla. Otto Harrassowitz. find
4. Arnold, W. (2011). Western Neo-Aramaic. In S. Weninger, G. Khan, M. P. Streck, & J. C. E. Watson (Eds.), The Semitic languages. An international handbook (pp. 685– 696). De Gruyter Mouton. find
5. Arnold, W. (2019). Das Neuwestaramäische: VI. Wörterbuch. Harrassowitz. find
6. Bergsträsser, G. (1915). Neuaramäische Märchen und andere Texte aus Maʿlūla. F.A. Brockhaus. find
7. Bergsträsser, G. (1933). Phonogramme im neuaramäischen Dialekt von Malula. Satzdruck und Satzmelodie. Verlag der Bayerischen Akademie der Wissenschaften. find
8. Boersma, P., & Weenink, D. J. (2021). Praat. Doing phonetics by computer [Computer program] (Version 6.1.49) [Computer software]. Version 6.1.49, retrieved from http://www.praat.org/ ↗
9. Eberhard, D. M., Simons, G. F., & Fennig, C. D. (Eds.). (2021). Ethnologue: Languages of the world (24th ed.). SIL International. Online version: http://www.ethnologue.com ↗
11. Eid, G., & Plag, I. (2021). Syllable structure and syllabification in Maaloula Aramaic. Submitted. find
12. Garofolo, J. S., Lamel, L. F., Fisher, W. M., Fiscus, J. G., Pallett, D. S., Dahlgren, N. L., & Zue, V. (1993). TIMIT Acoustic-Phonetic Continuous Speech Corpus LDC93S1. Linguistic Data Consortium. find
13. Godfrey, J. J., & Holliman, E. C. (1993). Switchboard-1 Release 2 LDC97S62. Linguistic Data Consortium. find
14. Godfrey, J. J., Holliman, E. C., & McDaniel, J. (1992). SWITCHBOARD: Telephone speech corpus for research and development. In Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (ICASSP-92) (pp. 517–520). San Francisco, CA. ↗
15. Heinrichs, W. (1990). Introduction. In W. Heinrichs (Ed.), Studies in Neo-Aramaic (pp. ix–xvii). Scholars Press. find
16. Khan, G., & Noorlander, P. M. (Eds.). (2021). Studies in the grammar and lexicon of Neo-Aramaic. Open Book Publishers. https://doi.org/10.11647/OBP.0209 ↗ our card
17. Kisler, T., Reichel, U. D., & Schiel, F. (2017). Multilingual processing of speech via web services. Computer Speech & Language, 45(September 2017), 326–347. ↗
18. McEnery, T., Xiao, R., & Tono, Y. (2006). Corpus-based language studies: An advanced resource book. Routledge. find
19. Moseley, C. (Ed.). (2010). Atlas of the World’s Languages in Danger. (3rd ed.). UNESCO Publishing. Online version: http://www.unesco.org/languages-atlas/ ↗
20. Pitt, M. A., Dilley, L., Johnson, K., Kiesling, S., Raymond, W., Hume, E., & Fosler-Lussier, E. (2007). Buckeye Corpus of Conversational Speech (2nd release). Department of Psychology, Ohio State University (Distributor). [www.buckeyecorpus.osu.edu] find
21. R Core Team. (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. ↗
22. Reich, S. (1937). Études sur les villages araméens de l’Anti-Liban. Institut Français de Damas. find
23. Schiel, F. (1999). Automatic phonetic transcription of non-prompted speech. In Proceedings of the ICPhS 1999 (pp. 607–610). San Francisco, August. find
24. Schiel, F. (2015). A statistical model for predicting pronunciation. In Proceedings of the ICPhS 2015. Glasgow, UK, paper 195. find
25. Spitaler, A. (1957). Neue Materialien zum aramäischen Dialekt von Maʿlūla. Zeitschrift Der Deutschen Morgenländischen Gesellschaft, 107(2), 299–339. 9. find
26. Language Resource References Arnold, W. (2003). Semitisches Tonarchiv (SemArch). Heidelberg University. Available online at http://semarch.ub.uni-heidelberg.de/ ↗
27. Eid, G., Seyffarth, E., Rihan, E., Arnold, W., & Plag, I. (2022). The Maaloula Aramaic Speech Corpus (MASC) (v1.0). Heinrich-Heine-Universität Düsseldorf. Zenodo. https://doi.org/10.5281/zenodo.6496714 ↗

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/lingbulk-2022-eid-maaloula-aramaic-speech-corpus-masc/