Elamur Research⌕
☰
home / papers / 2021 / The Articulatory and acoustics Effects of Pharyngeal Consonants on Adjacent Vowels in Arabic Language

The Articulatory and acoustics Effects of Pharyngeal Consonants on Adjacent Vowels in Arabic Language

Paper · 2021 · Fazia Karaoui, Amar Djeradi and Yves Laprie

Summary

The paper investigates how the Arabic pharyngeal consonants /ħ/ and /ʕ/ (written /è/ and /Q/ in the extracted text) are articulated and how they change neighbouring vowels in Moroccan Arabic. The authors focus on vertical larynx movement, which they say had not been measured explicitly for vowels in pharyngeal surroundings. They used cineradiographic (x-ray) film and audio of one native male speaker from the DOCVACIM database. Sentences were segmented and annotated in Praat, and each phone was aligned with its x-ray frames. Contours of the tongue, lips, jaw, palate, hyoid bone, larynx and glottis were traced to measure articulator displacement. Formants F1–F4 of the short vowels /a, i, u/ were then compared between pharyngeal and plain coronal contexts. In pharyngeal contexts the larynx rose by roughly 0.6–1.19 cm, and F1 increased by more than 200 Hz. The F2–F1 gap also narrowed, and the authors link this to a shorter vocal tract. The study reports a correlation between how high the larynx rises and how strong the coarticulation is. These measurements could help set the larynx-height parameter in articulatory models of speech.

Key points

  • Data come from DOCVACIM: 2700 vocal-tract x-ray images and 60 Moroccan Arabic sentences from one adult male speaker.
  • The analysis used 17 tokens of /ʕ/ and 8 of /ħ/. Vowels after or before pharyngeals totalled 15 tokens (10 /a/, 3 /i/, 2 /u/). The plain coronal baseline had 22 /a/, 20 /u/ and 13 /i/ tokens.
  • Mean larynx-centre raising was 0.872 cm for /ʕ/ and 1.644 cm for /ħ/. Mean hyoid raising was 0.724 cm for /ʕ/ and 0.597 cm for /ħ/.
  • In pharyngeal contexts, F1 rose by about 223 Hz for /a/, 214 Hz for /i/ and 239 Hz for /u/.
  • The F2–F1 distance shrank from about 1000 to about 800 Hz for /a/, and from about 700 to about 400 Hz for /u/. Voiceless /ħ/ generally had a stronger effect than voiced /ʕ/.

Interesting ideas

  • I1. Each acoustically segmented phone is time-aligned with its matching x-ray frames, so every acoustic label is backed by physical evidence from a second modality.
  • I2. Contour tracing is split by difficulty: rigid bones are tracked automatically, slightly overlapping organs semi-automatically, and only the tongue is drawn by hand. Human effort goes only where automation fails.
  • I3. Effects are always measured against a plain-consonant baseline rather than as absolute values. This separates the effect of the context from the vowel's own default properties.
  • I4. The strength of coarticulation is reduced to one interpretable number: the difference between F2–F1 in the baseline and in the target context. A value near zero means both contexts behave alike.
  • I5. The authors treat influence between neighbouring segments as running in both directions and as overlapping in time, not as a strictly one-way effect.

For Mova

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

  • I1 (frames): Each frame or semantic role could be stored with a pointer to the exact tree nodes and token span that justify it, which keeps every step traceable.
  • I2 (learn): Silver data could be routed the same way, with easy cases agreed by the witness parsers, uncertain ones sent to LLM teacher tiers, and only the hardest left for manual review.
  • I3 (learn): Evaluations could routinely report scores on a matched "plain" control set next to the target phenomenon, as a built-in negative control.
  • I4 (learn): A similar difference-of-gaps metric, such as the parser margin on a construction versus its plain counterpart, could flag constructions where Mova behaves unusually.
  • I5 (tokens): Tagging and lemma decisions could use features from both left and right neighbours, with explicit rules for when a neighbour overrides a token's default reading.

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

 Fields, topics and concepts

📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 works

🧭 Topics 2Research subjects and languages the work is about.Semitic languages · 3,066 worksSpeech & spoken language · 519 works

Cite

Cite as Fazia Karaoui, Amar Djeradi and Yves Laprie. 2021. The Articulatory and acoustics Effects of Pharyngeal Consonants on Adjacent Vowels in Arabic Language. Proceedings of the 4th International Conference on Natural Language and Speech Processing (ICNLSP 2021).

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

@inproceedings{karaoui2021articulatory,
  title = {{The Articulatory and acoustics Effects of Pharyngeal Consonants on Adjacent Vowels in Arabic Language}},
  author = {Fazia Karaoui, Amar Djeradi and Yves Laprie},
  booktitle = {Proceedings of the 4th International Conference on Natural Language and Speech Processing (ICNLSP 2021)},
  year = {2021},
  url = {https://aclanthology.org/2021.icnlsp-1.32/}
}

References

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

  1. Al-Ani. 1970. Arabic phonology an acoustical and physiological investigation. find
  2. Al-Tamimi. 2009. Effect of pharyngealisation on vowels revisited: Static and dynamic analyses of vowels in moroccan and jordanian arabic. In Workshop Pharyngeals and Pharyngealisation, Newcastle University. find
  3. Alwan. 1989. Perceptual cues for place of articulation for the voiced pharyngeal and uvular consonants. Journal of the Acoustical Society of America, 86(2):549–556. find
  4. Bin-Muqbil. 2006. Phonetic and phonological aspects of Arabic emphatics and gutturals, PhD dissertation, University of Wisconsin-Maddison. find
  5. Bouarourou. 2014. la gémination en tarifit, considérations phonologiques, étude acoustique et articulatoire, PhD dissertation,Strasbourg university. find
  6. Kusay Butcher. 1987. Some acoustic and aerodynamic consequences of pharyngeal consonants in iraqi arabic. phonetica, 44:156–172. find
  7. Byrd. 1994. Articulatory Timing in English Consonant Sequences, PhD dissertation, University of California, Los Angeles. find
  8. Byrd. 1996. Influences on articulatory timing in consonant sequences. Journal of Phonetics, 24:209–244. find
  9. Delattre. 1971. Pharyngeal feature in the consonants of arabic, german, french and american english. phonetica, 23:129–155. find
  10. Djeradi. 2006. Analyse et characterisation des fricatives d’arriére de l’arabe: au plan acoustique, articulatoire et “timing”,PhD dissertation, University of science and technology houari boumediene of Algiers. find
  11. Guilleminot Al-Maqtari Embarki, Yeou. 2007. An acoustic study of coarticulation in modern standard arabic and dialectal arabic: Pharyngealized vs. non-pharyngealized articulation. In 2007 In Proceedings of the 16th International Congress of Phonetic Sciences, pages 141–146. find
  12. Esling. 1996. Pharyngeal consonants and the aryepiglottic sphincter. Journal of the International Phonetic Association, 26(2):65–88. find
  13. Barkat Ghazali, Hamdi. 2002. Speech rhythm variation in arabic dialects. find
  14. Ghazeli. 1977. Back Consonants and Backing Coarticulation in Arabic,PhD dissertation,University of Texas, Austin. find
  15. Vaxelaire Elie Laprie, Sock. 2014. Comment faire parler les images aux rayons x du conduit vocal. In In SHS Web of Conferences, volume 8, page 14. Association for Computational Linguistics. find
  16. Baer Laufer. 1988. The emphatic and pharyngeal sounds in hebrew and in arabic. language and Speech, 31(2):181–205. find
  17. Marchal. 2010. From Speech Physiology to Linguistic Phonetics,John Wiley and Sons,ISBN, 0470610409, 978047610404. find
  18. McCarthy. 1994. The phonetics and phonology of semitic pharyngeals. In in Patricia A. Keating, ed., Phonological Structure and Phonetic Form, Papers in Laboratory Phonology Cambridge, UK: Cambridge University Press, page 191–233. find
  19. Laprie Perrier-Vaxelaire Sock, Hirsch. 2011. An xray database, tools and procedures for the study of speech production. In 9th International Seminar on Speech Production (ISSP 2011), Montreal, Canada, pages 41–48. find
  20. Sundburg. 2003. Research on the singing voice in retrospect. In Speech Music and Hearing Laboratory-Quarterly Progress and Status Report, Stockholm, pages 11–22. find
  21. Sylak. 2013. Pharyngealization in chechen is guttural-ization. In Proceedings of the 37th Annual Meeting of the Berkeley Linguistics Society: Special Session on Languages of the Caucasus, pages 81–95. find
  22. Zawaydeh. 1999. The phonetics and phonology of gutturals in Arabic, Ph.D. Dissertation, Indiana University. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/karaoui-2021-articulatory-acoustics-effects-pharyngeal-consonants/