Elamur Research⌕
☰
home / papers / 2026 / A Database of Romance Clitics With Speech Samples

A Database of Romance Clitics With Speech Samples

Summary

Most Romance languages have unstressed clitic pronouns. However, several minority varieties in southern France, northern and southern Italy, Corsica, Sardinia and the Balearic Islands show clitics that carry stress or shift it within the verb+clitic unit. These varieties are endangered, poorly documented and are losing ground to the national standard languages. The authors present the Romance Clitics Database (RoCDat) to document this pattern. It combines audio recordings, transcriptions and detailed linguistic coding.

Between 2007 and 2010, more than 75 native speakers were recruited through local contacts. Each translated about 100 verb+clitic phrases from a standard language (Italian, Spanish, Catalan or French) into their own variety. The phrases covered full verb paradigms and the possible clitic combinations. Each utterance was annotated for 42 phonological, morphological and syntactic features, along with anonymized sociolinguistic metadata.

The resource is a per-utterance CSV with a searchable web interface, and the interface code can be reused for similar projects. The authors plan to release both publicly. RoCDat supports linguistic research on how morphology, syntax and prosody interact, and it gives low-resource NLP work a small, targeted dataset.

Key points

  • Coverage: RoCDat spans nine regions: the French department of Pyrénées-Atlantiques; the islands of Corsica and Sardinia; the Italian regions of Liguria, Campania and Basilicata; and the Balearic islands of Mallorca, Menorca and Formentera.
  • Size: It contains about 266 minutes of speech from 75 speakers (36 women, 39 men).
    • Italy: 35 speakers, about 142.5 minutes.
    • Spain: 28 speakers, about 59.5 minutes.
    • France: 12 speakers, about 64.2 minutes.
    • Sardinia alone contributes 24 speakers and about 122 minutes.
  • Annotation: The 42 coded features include:
    • stress position (final, penultimate, antepenultimate, pre-antepenultimate);
    • gemination;
    • clitic placement;
    • verb form features;
    • pronoun role, form, gender, number and person.
  • Clitic placement: In single-verb utterances, enclisis accounts for about 68% of cases and proclisis for about 32%.
  • Comparison with related resources: The authors find no directly comparable dataset. Related resources include C-ORAL-ROM, Rhapsodie, COSER, ALPI, AnCora and the Italian Stanford Dependency Treebank, but none combines clitic-focused annotation with speech from these varieties.

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: Linguistics & languages

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.63317/5azno6dg9d4p

@inproceedings{qaddoumi2026database,
  title = {{A Database of Romance Clitics With Speech Samples}},
  author = {Qaddoumi, Abdelrahim and Rambow, Owen and Repetti, Lori and Ordóñez, Francisco},
  booktitle = {Proceedings of the SIGUL 2026 Joint Workshop with ELE, EURALI, and DCLRL: Towards Inclusivity and Equality: Language Resources and Technologies for Under-Resourced and Endangered Languages},
  year = {2026},
  doi = {10.63317/5azno6dg9d4p},
  url = {https://doi.org/10.63317/5azno6dg9d4p}
}

References

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

  1. Cristina Bosco, Simonetta Montemagni, Maria Simi, et al. 2013. Converting italian treebanks: Towards an italian stanford dependency treebank. In Proceedings of the 7th Linguistic Annotation Workshop and Interoperability with Discourse, pages 61–69. The Association for Computational Linguistics. find
  2. To add audio, users need to create an S3 URL path for utterances with audio data. Then the configuration requires adding the DATABASE URL, S3 bucket, and Allowed Hosts. The repository is shipped with Docker/Gunicorn/Whitenoise for one-command deployment and static asset handling. We will include a G… find
  3. Philippe Boula de Mareüil, Frédéric Vernier, and Albert Rilliard. 2017. Enregistrements et tran-Pronoun 1 Pronoun 2 Pronoun 3 1614 1352 13 find
  4. Masculine Feminine n % n % 1116 958 12 67.35 70.23 92.31 498 394 1 30.05 28.89 7.69 *Table 9: Pronoun gender distributions.* Source Pronoun 1 Pronoun 2 Pronoun 3 Count 3769 1598 16 DO IO Subject n % n % n % 1820 1326 10 48.19 82.57 62.50 1843 271 6 48.80 16.87 37.50 106 1 – 2.81 0.06 – *Table 10: Pr… find
  5. Atanas Tchobanov. 2014. Rhapsodie: a prosodic-syntactic treebank for spoken french. In Language Resources and Evaluation Conference. find
  6. Massimo Cerruti and Riccardo Regis. 2014. Standardization patterns and dialect/standard convergence: A northwestern italian perspective. Language in society, 43(1):83–111. find
  7. Rosangela Lai. 2017. Stress shift under cliticization in nuorese sardinian. Quaderni di linguistica e studi orientali, 3:183–199. find
  8. Luis Chiruzzo and Dina Wonsever. 2018. Spanish hpsg treebank based on the ancora corpus. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018). find
  9. John M Levis. 2018. Intelligibility, oral communication, and the teaching of pronunciation. Cambridge University Press. find
  10. Francisco Ordóñez and Lori Repetti. 2006. Stressed enclitics? AMSTERDAM STUDIES IN THE THEORY AND HISTORY OF LINGUISTIC SCIENCE SERIES 4, 276:167. find
  11. Emanuela Cresti, Fernanda Bacelar Do Nascimento, Antonio Moreno-Sandoval, Jean Veronis, Philippe Martin, and Khalid Choukri. 2004. The c-oral-rom corpus. a multilingual resource of spontaneous speech for romance languages. In LREC. find
  12. Alan Ramponi and Camilla Casula. 2023. Diatopit: A corpus of social media posts for the study of diatopic language variation in italy. In Tenth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial 2023), pages 187–199. find
  13. Alexander Erdmann and Nizar Habash. 2018. Complementary strategies for low resourced morphological modeling. In Proceedings of the Fifteenth Workshop on Computational Research in Phonetics, Phonology, and Morphology, pages 54–65. find
  14. Lori Repetti. 2018. Fieldwork and building corpora for endangered varieties. Ayres-Bennet, W. & J. Carruthers (edd.): Manual of Romance Sociolinguistics. De Gruyter, Berlin/Boston, pages 114– 133. find
  15. Inés Fernández-Ordóñez. 2005. Corpus oral y sonoro del español rural. Madrid, Universidad Autónoma de Madrid. find
  16. Mariona Taulé, Maria Antònia Martí, and Marta Recasens. 2008. Ancora: Multilevel annotated corpora for catalan and spanish. In Lrec, volume 2008, pages 96–101. find
  17. Pilar García Mouton. 2016. Inés fernández-ordóñez, david heap. Maria Pilar PEREA, João SARAMAGO & Xulio SOUSA. find
  18. Francis Zheng, Edison Marrese-Taylor, and Yutaka Matsuo. 2024. Improving low-resource machine translation for formosan languages using bilingual lexical resources. In Findings of the Association for Computational Linguistics: ACL 2024, pages 11248–11259. find
  19. Robert A Hall Jr. 1942. Sprach-und sachatlas italiens und der südschweiz. find
  20. Anne Lacheret, Sylvain Kahane, Julie Beliao, Anne Dister, Kim Gerdes, Jean-Philippe Goldman, Nicolas Obin, Paola Pietrandrea, and -l-Pronoun 1 Pronoun 2 Pronoun 3 2615 1487 12 Reflexive Possessive Locative Partitive Impersonal Other 1430 1315 9 54.68 88.43 75.00 613 48 – 23.44 3.23 – 206 – – 7.88 –… find
  21. Bosco, Cristina and Montemagni, Simonetta and Simi, Maria and others. 2014–present. Converting italian treebanks: Towards an italian stanford dependency treebank. Stanford deps/UD-converted Italian treebank; Evalita 2014 origin. find
  22. Graziano Tisato. 2009. Ais digital atlas and navigation software. Online resource (no audio). Accessed 24 Oct 2025. find
  23. Boula de Mareüil, Philippe and Vernier, Frédéric and Rilliard, Albert. 2017–. Enregistrements et transcriptions pour un atlas sonore des langues régionales de France. Interactive recordings+transcriptions of regional languages of France. find
  24. Cresti, Emanuela and Do Nascimento, Fernanda Bacelar and Moreno-Sandoval, Antonio and Veronis, Jean and Martin, Philippe and Choukri, Khalid. 2004. The C-ORAL-ROM CORPUS. A Multilingual Resource of Spontaneous Speech for Romance Languages. ELRA. Multilingual spontaneous speech (IT/FR/PT/ES); LREC 20… find
  25. Fernández-Ordóñez, Inés. 2005–present. Corpus oral y sonoro del español rural. Searchable audio+transcripts of rural Spanish; ongoing releases. find
  26. Lacheret, Anne and Kahane, Sylvain and Beliao, Julie and Dister, Anne and Gerdes, Kim and Goldman, Jean-Philippe and Obin, Nicolas and Pietrandrea, Paola and Tchobanov, Atanas. 2014. Rhapsodie: a prosodic-syntactic treebank for spoken french. 57 five-minute samples, orthographic+phonetic alignment;… find
  27. Pilar Prieto, Rafèu Sichel-Bazin, and Trudel (eds.) Meisenburg. 2007–2014. Atlàs interactiu de l’intonacion de l’occitan. Online resource. Accessed 24 Oct 2025. find
  28. Feature Two-verb clitic position Proclisis Enclisis Verb-cl-verb Other-pattern n % n % n % n % 52.68 22.53 22.17 2.62 *Table 13: Two-verb clitic position.* Feature Count Three-verb clitic position Proclisis V-V-cl-V Enclisis V-cl-V-V Other-pattern n % n % n % n % n % 42.40 25.60 21.60 6.40 4.00 *Tab… find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/qaddoumi-2026-database-romance-clitics-speech-samples/