Final published version, 219 KB, PDF document
Available under license: CC BY: Creative Commons Attribution 4.0 International License
Final published version
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper
}
TY - GEN
T1 - Dimensions of structure and variability in the human vocal tract
AU - Vaughan-Williams, Katherine
AU - Moran, Steven
AU - Kirkham, Sam
N1 - Conference code: 13th
PY - 2024/5/13
Y1 - 2024/5/13
N2 - A defining characteristic of the human vocal tract is a complex dynamic between structure and variability. Across a population we observe considerable variability in vocal tract dimensions, but variation in one dimension is rarely independent of other dimensions. Are some of these relationships more variable than others, or do there exist invariants in the morphology of the vocal tract? In this study, we report a data-driven investigation into the relationship between vocal tract dimensions based on multi-speaker real-time magnetic resonance imaging data. We discover different sub-populations in the data, which correspond to groups of speakers that share a common relationship between vocal tract parameters. This suggests a range of complex patterns of co-variation in the morphology of the human vocal tract. We conclude by speculating on the possible implications of these results for understanding individual differences in speech production.
AB - A defining characteristic of the human vocal tract is a complex dynamic between structure and variability. Across a population we observe considerable variability in vocal tract dimensions, but variation in one dimension is rarely independent of other dimensions. Are some of these relationships more variable than others, or do there exist invariants in the morphology of the vocal tract? In this study, we report a data-driven investigation into the relationship between vocal tract dimensions based on multi-speaker real-time magnetic resonance imaging data. We discover different sub-populations in the data, which correspond to groups of speakers that share a common relationship between vocal tract parameters. This suggests a range of complex patterns of co-variation in the morphology of the human vocal tract. We conclude by speculating on the possible implications of these results for understanding individual differences in speech production.
U2 - 10.21437/issp.2024-69
DO - 10.21437/issp.2024-69
M3 - Conference contribution/Paper
SP - 274
EP - 277
BT - Proceedings of the 13th International Seminar on Speech Production
A2 - Fougeron, Cécile
A2 - Perrier, Pascal
PB - International Speech Communication Association
T2 - ISSP 2024 : 13th International Seminar on Speech Production
Y2 - 13 May 2024 through 17 May 2024
ER -