Unifying Amplitude and Phase Analysis: A Compositional Data Approach to Functional Multivariate Mixed-Effects Modeling of Mandarin Chinese

Overview

Journal J Am Stat Assoc

Specialty Public Health

Date 2015 Dec 23

PMID 26692591

Citations 6

Authors

P Z Hadjipantelis

J A D Aston

H G Muller

J P Evans

Affiliations

Soon will be listed here.

Abstract

Mandarin Chinese is characterized by being a tonal language; the pitch (or ) of its utterances carries considerable linguistic information. However, speech samples from different individuals are subject to changes in amplitude and phase, which must be accounted for in any analysis that attempts to provide a linguistically meaningful description of the language. A joint model for amplitude, phase, and duration is presented, which combines elements from functional data analysis, compositional data analysis, and linear mixed effects models. By decomposing functions via a functional principal component analysis, and connecting registration functions to compositional data analysis, a joint multivariate mixed effect model can be formulated, which gives insights into the relationship between the different modes of variation as well as their dependence on linguistic and nonlinguistic covariates. The model is applied to the COSPRO-1 dataset, a comprehensive database of spoken Taiwanese Mandarin, containing approximately 50,000 phonetically diverse sample contours (syllables), and reveals that phonetic information is jointly carried by both amplitude and phase variation. Supplementary materials for this article are available online.

Citing Articles

Functional regression clustering with multiple functional gene expressions.

Conde S, Tavakoli S, Ezer D PLoS One. 2024; 19(11):e0310991.

PMID: 39585813 PMC: 11588248. DOI: 10.1371/journal.pone.0310991.

Regression and alignment for functional data and network topology.

Tu D, Wrobel J, Satterthwaite T, Goldsmith J, Gur R, Gur R Biostatistics. 2024; 26(1).

PMID: 39140988 PMC: 11822954. DOI: 10.1093/biostatistics/kxae026.

Trajectories of brain volumes in young children are associated with maternal education.

Zhu C, Chen Y, Muller H, Wang J, OMuircheartaigh J, Bruchhage M Hum Brain Mapp. 2023; 44(8):3168-3179.

PMID: 36896867 PMC: 10171562. DOI: 10.1002/hbm.26271.

Registration for exponential family functional data.

Wrobel J, Zipunnikov V, Schrack J, Goldsmith J Biometrics. 2018; 75(1):48-57.

PMID: 30129091 PMC: 10585654. DOI: 10.1111/biom.12963.

Novel metrics for growth model selection.

Grigsby M, Di J, Leroux A, Zipunnikov V, Xiao L, Crainiceanu C Emerg Themes Epidemiol. 2018; 15:4.

PMID: 29483933 PMC: 5824542. DOI: 10.1186/s12982-018-0072-z.

References

Jacewicz E, Fox R, Wei L . Between-speaker and within-speaker variation in speech tempo of American English. J Acoust Soc Am. 2010; 128(2):839-50. PMC: 2933259. DOI: 10.1121/1.3459842. View

Evans J, Chu M, Aston J, Su C . Linguistic and human effects on F(0) in a tonal dialect of Qiang. Phonetica. 2010; 67(1-2):82-99. DOI: 10.1159/000319380. View

Lindquist M, Spicer J, Asllani I, Wager T . Estimating and testing variance components in a multi-level GLM. Neuroimage. 2011; 59(1):490-501. PMC: 3195889. DOI: 10.1016/j.neuroimage.2011.07.077. View

Schielzeth H, Forstmeier W . Conclusions beyond support: overconfident estimates in mixed models. Behav Ecol. 2009; 20(2):416-420. PMC: 2657178. DOI: 10.1093/beheco/arn145. View

Bruns A . Fourier-, Hilbert- and wavelet-based signal analysis: are they really different approaches?. J Neurosci Methods. 2004; 137(2):321-32. DOI: 10.1016/j.jneumeth.2004.03.002. View

Laird N, Ware J . Random-effects models for longitudinal data. Biometrics. 1982; 38(4):963-74. View

Tang R, Muller H . Time-synchronized clustering of gene expression trajectories. Biostatistics. 2008; 10(1):32-45. DOI: 10.1093/biostatistics/kxn011. View

Cangelosi R, Goriely A . Component retention in principal component analysis with application to cDNA microarray data. Biol Direct. 2007; 2:2. PMC: 1797006. DOI: 10.1186/1745-6150-2-2. View

Guo W . Functional mixed effects models. Biometrics. 2002; 58(1):121-8. DOI: 10.1111/j.0006-341x.2002.00121.x. View

10.

Grabe E, Kochanski G, Coleman J . Connecting intonation labels to mathematical descriptions of fundamental frequency. Lang Speech. 2007; 50(Pt 3):281-310. DOI: 10.1177/00238309070500030101. View

11.

Prom-On S, Xu Y, Thipakorn B . Modeling tone and intonation in Mandarin and English as a process of target approximation. J Acoust Soc Am. 2009; 125(1):405-24. DOI: 10.1121/1.3037222. View

12.

Hadjipantelis P, Aston J, Evans J . Characterizing fundamental frequency in Mandarin: a functional principal component approach utilizing mixed effect models. J Acoust Soc Am. 2012; 131(6):4651-64. DOI: 10.1121/1.4714345. View

13.

Lee S, Byrd D, Krivokapic J . Functional data analysis of prosodic effects on articulatory timing. J Acoust Soc Am. 2006; 119(3):1666-71. PMC: 2392894. DOI: 10.1121/1.2161436. View

14.

Ramsay J, Munhall K, Gracco V, Ostry D . Functional data analyses of lip motion. J Acoust Soc Am. 1996; 99(6):3718-27. DOI: 10.1121/1.414986. View

15.

Friedman J, Hastie T, Tibshirani R . Regularization Paths for Generalized Linear Models via Coordinate Descent. J Stat Softw. 2010; 33(1):1-22. PMC: 2929880. View

16.

Barr D, Levy R, Scheepers C, Tily H . Random effects structure for confirmatory hypothesis testing: Keep it maximal. J Mem Lang. 2014; 68(3). PMC: 3881361. DOI: 10.1016/j.jml.2012.11.001. View

17.

Taylor P . Analysis and synthesis of intonation using the Tilt model. J Acoust Soc Am. 2000; 107(3):1697-714. DOI: 10.1121/1.428453. View

18.

Hyvarinen A, Oja E . Independent component analysis: algorithms and applications. Neural Netw. 2000; 13(4-5):411-30. DOI: 10.1016/s0893-6080(00)00026-5. View

19.

Koenig L, Lucero J, Perlman E . Speech production variability in fricatives of children and adults: results of functional data analysis. J Acoust Soc Am. 2008; 124(5):3158-70. PMC: 2677351. DOI: 10.1121/1.2981639. View