The prosodic structure in French: Analysis with deep learning networks
DOI:
https://doi.org/10.36505/ExLing-2021/0040/000513Keywords:
Deep learning, French, ToBI, prosodic structureAbstract
Experiments using supervised deep learning algorithms were executed on a large set of sentences selected from the SIWIS (Yamagishi et al., 2016) corpus in order to investigate properties of the prosodic structure in French. Based on a phonological model assuming dependency relations between accent phrases (Martin, 2018), the training corpus was annotated with a modified ToBI notation system, encoding prosodic events located on stressed syllables (vowels), i.e. pitch accents, and syntactic groups final syllables (vowels), i.e. boundary tones, which in French occur on the same accent phrases final syllable.References
Martin, Ph. 2018. Intonation, structure prosodique et ondes cérébrales, London: ISTE, 322 p.
Rossi, M. 1971. Le seuil de glissando ou seuil de perception des variations tonales pour la parole. Phonetica. n° 23, 1-33.
Yamagishi, J. et al. 2016. The SIWIS French Speech Synthesis Database, University of Edinburgh. School of Informatics. The Centre for Speech Technology Research, https://doi.org/10.7488/ds/1705.
Downloads
Published
01-01-2021
Issue
Section
Proceedings Papers
License
Articles are published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
How to Cite
The prosodic structure in French: Analysis with deep learning networks. (2021). Linguistic Proceedings Series, 12(1), 157-160. https://doi.org/10.36505/ExLing-2021/0040/000513
