Elamur Research⌕
☰
home / papers / 2026 / SOMVOICE: A First Dataset to Study the Effects of Sleep Deprivation on Voice Characteristics of Healthy French Speakers

SOMVOICE: A First Dataset to Study the Effects of Sleep Deprivation on Voice Characteristics of Healthy French Speakers

Summary

Automatic voice analysis could help clinicians monitor excessive sleepiness outside the clinic. This condition is widespread and linked to many disorders. Existing speech corpora fall short in different ways: some target task performance rather than sleepiness, some use annotation scales not accepted in sleep medicine, and others record patients rather than healthy people.

The authors present SOMVOICE, a French read-speech corpus. Each of 28 healthy native speakers was recorded in a sleep laboratory twice, once after ordinary rest and once after a full night without sleep, in randomised order. Candidates were screened medically and, for the first time in such a resource, for reading skill. Each recording is paired with an objective clinical sleepiness measure and with self-reports of sleepiness, fatigue and anxiety.

A preliminary analysis with linear mixed models shows that sleep loss clearly raises objective and subjective sleepiness and fatigue but has a weaker effect on anxiety. It also makes participants read more slowly and pause more. The corpus offers a controlled within-speaker contrast for developing French voice biomarkers of sleepiness.

Key points

  • Participants and recruitment: 28 participants (12 men, 16 women, aged 18–50) were included between July 2022 and June 2024. Each received 400€ for completing four visits. The protocol is registered as NCT04942574 on ClinicalTrials.gov.
  • Medical screening: Candidates had a clinical interview to rule out sleep, mood and other relevant disorders. This was followed by seven nights of actimetry and ambulatory polygraphy, with sleep efficiency required to be at least 85% and sleep apnea and restless legs syndrome excluded.
  • Reading screening: Reading ability was tested with tasks from the ECLA16+ battery, including reading the text L'Alouette, word and pseudo-word reading, pseudo-word repetition, and symbol crossing. Candidates had to score above the 25th percentile of norms established on 311 French 16-year-olds.
  • Effects of sleep deprivation: Sleep loss strongly affected objective and subjective sleepiness and fatigue, with a smaller effect on anxiety. Speech rate dropped (longer recordings, lower phoneme rate), and both the number of pauses and the pause ratio increased.
  • Speech-processing pipeline: Phoneme segmentation combined a conformer ASR trained on ESTER (WER 14.6%, CER 5.7%) with the Montreal Forced Aligner. Evaluated on the planned subcorpus of Rhapsodie, the complete pipeline achieved an F-score of 0.72, with errors on 10.1% of phonemes (Phoneme Error Rate).

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 10.63317/5gzhbud3vbof

@inproceedings{martin2026somvoice,
  title = {{SOMVOICE: A First Dataset to Study the Effects of Sleep Deprivation on Voice Characteristics of Healthy French Speakers}},
  author = {Martin, Vincent P. and Rouas, Jean-Luc and Beaumard, Colleen and Philip, Pierre},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/5gzhbud3vbof},
  url = {https://doi.org/10.63317/5gzhbud3vbof}
}

References

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

  1. Torbjorn Åkerstedt and Mats Gillberg. 1990. Subjective and objective sleepiness in the active individual. Int J Neurosci, 52:29–37. find
  2. S. Boyer, R. El-Yagoubi, M. Tiberge, R. Ruiz, and A. Daurat. 2016. Paramètres Acoustiques de la Voix et Privation de Sommeil. In CFA/VISHNO. find
  3. Lakshmi Swathi Dhupati, Sibsambhu Kar, Aparna Rajaguru, and Aurobinda Routray. 2010. A novel drowiness detection scheme based on speech analysis with validation using simultaneous EEG recordings. In IEEE - Int. CASE, pages 917–921. find
  4. Sylvain Galliano, Guillaume Gravier, and Laura Chaubard. 2009. The ESTER 2 Evaluation Campaign for the Rich Transcription of French Radio Broadcasts. In Interspeech 2009, pages 2583– 2586. find
  5. John Goddard, George Tay, Jennifer Fry, Mark Davis, Deanne Curtin, and Irene Szollosi. 2021. Multiple Sleep Latency Test: When are 4 naps enough? Journal of Clinical Sleep Medicine, 17(3):491–497. find
  6. Martin Golz, David Sommer, Mo Chen, Dalino Mandic, and Udo Trutschel. 2007. Feature Fusion for the Detection of Microsleep Events. Journal of VLSI Signal Processing, 49:329–342. find
  7. Maki Jike, Osamu Itani, Norio Watanabe, Daniel J. Buysse, and Yoshitaka Kaneita. 2018. Long sleep duration and health outcomes: A systematic review, meta-analysis and meta-regression. Sleep Medicine Reviews, 39:25–36. find
  8. Bhanu Prakash Kolla, Jian-Ping He, Meghna P. Mansukhani, Mark A. Frye, and Kathleen Merikangas. 2020. Excessive sleepiness and associated symptoms in the U.S. adult population: Prevalence, correlates, and comorbidity. Sleep Health, 6(1):79–87. find
  9. Lois E. Krahn, Donna L. Arand, Alon Y. Avidan, David G. Davila, William A. DeBassio, Chad M. Ruoff, and Christopher G. Harrod. 2021. Recommended protocols for the Multiple Sleep Latency Test and Maintenance of Wakefulness Test in adults: Guidance from the American Academy of Sleep Medicine. Journal… find
  10. Anne Lacheret-Dujour, Sylvain Kahane, and Paola Pietrandrea, editors. 2019. Rhapsodie: A Prosodic and Syntactic Treebank for Spoken French, volume 89 of Studies in Corpus Linguistics. John Benjamins Publishing Company, Amsterdam. find
  11. Kathryn A Lee, Gregory Hicks, and German Nino-Murcia. 1991. Validity and reliability of a scale to assess fatigue. Psychiatry research, 36(3):291– 298. find
  12. Vincent P. Martin, Brice Arnaud, Jean-Luc Rouas, and Pierre Philip. 2022. Does sleepiness influence reading pauses in hypersomniac patients? In Speech Prosody 2022, pages 62–66. ISCA. find
  13. Vincent P. Martin, Colleen Beaumard, Jean-Luc Rouas, and Yaru Wu. 2024a. Is automatic phoneme recognition suitable for speech analysis? Temporal and performance evaluation of an Automatic Speech Recognition model in spontaneous French. In Speech Prosody 2024, pages 1120–1124. find
  14. Vincent P. Martin, Aymeric Ferron, Jean-Luc Rouas, Takaaki Shochi, Lucile Dupuy, and Pierre Philip. 2023a. Physiological vs. Subjective sleepiness: What can human hearing estimate better? In International Conference on Phonetic Science (ICPhS) 2023, pages 196–200. find
  15. Vincent P. Martin, Régis Lopez, Yves Dauvilliers, Jean-Luc Rouas, Pierre Philip, and Jean-Arthur Micoulaud-Franchi. 2023b. Sleepiness in adults: An umbrella review of a complex construct. Sleep Medicine Reviews, 67:101718. find
  16. Vincent P. Martin, Jean-Luc Rouas, Jean-Arthur Micoulaud-Franchi, and Pierre Philip. 2020. find
  17. The Objective and Subjective Sleepiness Voice Corpora. In Language Resources and Evaluation Conference 2020, pages 6525–6533, Marseille, France. find
  18. Vincent P. Martin, Jean-Luc Rouas, Jean-Arthur Micoulaud-Franchi, Pierre Philip, and Jarek Krajewski. 2021. How to Design a Relevant Corpus for Sleepiness Detection Through Voice? Frontiers in Digital Health, 3:686068. find
  19. Vincent P. Martin, Jean-Luc Rouas, and Pierre Philip. 2024b. Automatic detection of sleepiness-related symptoms and syndromes using voice and speech biomarkers. Biomedical Signal Processing and Control, 91:105989. find
  20. Michael McAuliffe, Michaela Socolof, Sarah Mihuc, Michael Wagner, and Morgan Sonderegger. 2017. find
  21. Montreal Forced Aligner: Trainable Text-Speech Alignment Using Kaldi. In Interspeech 2017, pages 498–502. ISCA. find
  22. Vikram Ramanarayanan, Adam C. Lammert, Hannah P. Rowe, Thomas F. Quatieri, and Jordan R. Green. 2022. Speech as a Biomarker: Opportunities, Interpretability, and Challenges. Perspectives of the ASHA Special Interest Groups, 7(1):276–283. find
  23. Allan Rechtschaffen and A Kales. 1968. A Manual for Standardized Terminology, Techniques and Scoring System for Sleep Stages in Human Subjects, ucla edition. Los Angeles: BIS/BRI. find
  24. Björn Schuller, Anton Batliner, Christian Bergler, Florian B. Pokorny, Jarek Krajewski, Margaret Cychocz, Ralf Vollman, Sonja-Dana Roelen, Sebastian Schnieder, Elika Bergelson, Alejandrina Cristia, Amanda Seidl, Anne Warlaumont, Lisa Yankowitz, Elmar Nöth, Shahin Amiriparian, Simone Hantke, and Maxi… find
  25. The INTERSPEECH 2019 Computational Paralinguistics Challenge: Styrian Dialects, Continuous Sleepiness, Baby Sounds & Orca Activity. In Interspeech 2019. find
  26. Björn Schuller, Stefan Steidl, Anton Batliner, Florian Schiel, and Jarek Krajewski. 2011. find
  27. The IN-TERSPEECH 2011 Speaker State Challenge. In Interspeech 2011, pages 3201–3204. find
  28. Björn Schuller, Stefan Steidl, Anton Batliner, Florian Schiel, Jarek Krajewski, Felix Weninger, and Florian Eyben. 2013. Medium-term speaker states-A review on intoxication, sleepiness and the first challenge. Comput. Speech Lang., 28(2):346– 374. find
  29. Alexander J. Scott, Thomas L. Webb, Marrissa Martyn-St James, Georgina Rowse, and Scott Weich. 2021. Improving sleep quality leads to better mental health: A meta-analysis of randomised controlled trials. Sleep Medicine Reviews, 60:101556. find
  30. Jianhua Shen, Joseph Barbera, and Colin M. Shapiro. 2006. Distinguishing sleepiness and fatigue: Focus on definition and measurement. Sleep Medicine Reviews, 10(1):63–76. find
  31. Etienne Thoret, Thomas Andrillon, Caroline Gau-riau, Damien Léger, and Daniel Pressnitzer. 2024. Sleep deprivation detected by voice analysis. PLOS Computational Biology, 20(2):e1011849. find
  32. Bang Tran, Youxiang Zhu, Xiaohui Liang, James W. Schwoebel, and Lindsay A. Warrenburg. 2022. Speech Tasks Relevant to Sleepiness Determined With Deep Transfer Learning. In ICASSP 2022, pages 6937–6941. find
  33. Valerie S. L. Williams, Robert J. Morlock, and Douglas Feltner. 2010. Psychometric evaluation of a visual analog scale for the assessment of anxiety. Health and Quality of Life Outcomes, 8:57. find
  34. Dong Yang, Tomoki Koriyama, and Yuki Saito. 2024. Frame-Wise Breath Detection with Self-Training: An Exploration of Enhancing Breath Naturalness in Text-to-Speech. In Interspeech 2024, pages 4928–4932. ISCA. find

↑ back to top