Leveraging the temporal dynamics of anticipatory vowel-to-vowel coarticulation in linguistic prediction: A statistical modeling approach

Flego, Stefon M.; Forrest, Jon

doi:10.1016/j.wocn.2021.101093

Cited by 3 publications

(5 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent work using Bayesian multinomial regression models relaxes these assumptions, allowing for additional flexibility. For example, Flego and Forrest (2021) predict vowel identity from the acoustics of preceding phones, and are able to examine the effects of both individual speakers and preceding phones, via crossed random effects (see also Williams et al, 2019, Barreda and Silbert, 2023, Ch. 12, Moghiseh et al, 2023.…”

Section: Examples: Discrete Ymentioning

confidence: 99%

“…For example, Flego and Forrest (2021) and Kim and Tilsen (2022) assess the nature of the relationship between phonetic variables and a dependent variable (time and speech rate, respectively), comparing Bayesian regression models with different degrees of non-linearity using WAIC. Kim and Tilsen (2022) describe their study as exploratory, which is consistent with the Bayesian framework: the analyst can compare many models, to inform future confirmatory studies, without committing to one model being 'right' (see Sec.…”

Section: Model Evaluation and Comparisonmentioning

confidence: 99%

“…Fox and Jacewicz, 2009) and patterns of coarticulation (see e.g. Flego and Forrest, 2021). Functional PCA (Section 4.2.3) attenuates this issue by discovering prevalent dimensions of variation in curve shapes directly from the data, but FPCA still focuses on a few parameters and may miss subtle or local patterns.…”

Section: Modelling Parameters Of Curvesmentioning

confidence: 99%

“…coarticulation) are factored out, suggesting that human perception may also factor out context rather than classifying based on raw acoustic cues. Flego and Forrest (2021) extended this approach to formant dynamics using Bayesian regression models. Yu (2017) examined the role of temporal resolution in Cantonese tone perception by humans and machines, using SVMs, and found that just a few samples per syllable were enough for accurate classification.…”

Section: Human Versus Machine Comparisonmentioning

confidence: 99%

See 3 more Smart Citations

Advancements of Phonetics in the 21st century: Quantitative data analysis

Sonderegger,

Soskuthy

2024

Preprint

View full text Add to dashboard Cite

Phonetic research in the 21st century has relied heavily on quantitative analysis. This article reviews the evolu- tion of common practices and the emergence of newer techniques. Using a detailed literature survey, we show that most work follows a mainstream, which has shifted from ANOVAs to mixed-effects regression models over time. Alongside this mainstream, we highlight the increasing use of a diverse methodological toolbox, especially Bayesian methods and dynamic methods, for which we provide comprehensive reviews. Bayesian methods offer flexibility in model specification, interpretation, and incorporation of prior knowledge. Dynamic methods, such as GAMMs and functional data analysis, capture non-linear patterns in acoustic and articulatory data. Machine learning techniques, such as random forests, expand the questions and types of data phoneticians can analyze. We also discuss the grow- ing importance of open science practices promoting replicability and transparency. We argue that the future lies in a diverse methodological toolbox, with techniques chosen based on research questions and data structure.

show abstract

Section: Examples: Discrete Ymentioning

confidence: 99%

Section: Model Evaluation and Comparisonmentioning

confidence: 99%

Section: Modelling Parameters Of Curvesmentioning

confidence: 99%

Section: Human Versus Machine Comparisonmentioning

confidence: 99%

See 2 more Smart Citations

Advancements of Phonetics in the 21st century: Quantitative data analysis

Sonderegger,

Soskuthy

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…More recently Bayesian approaches that can directly test the hypothesis have been adopted into linguistic studies (Flego & Forrest, 2021;Levshina, 2016;MacKenzie, 2020;Nalborczyk et al, 2019;Vasishth, Chen, Li, & Guo, 2013 inter alia). One major advantage of taking a Bayesian approach is the ability to incorporate a researcher's prior knowledge into the model through the setting of priors.…”

Section: Bayesian Multinomial Modellingmentioning

confidence: 99%

Directional constructions in Matukar Panau

Davey,

Barth

2023

APLV

View full text Add to dashboard Cite

The Oceanic language Matukar Panau has three equivalent morphosyntactic strategies for describing the direction of the event represented by a verb, with a system of ten directional morphemes that can appear in each construction. This variation is explored using a corpus from Matukar Panau and analysed quantitatively with Bayesian regression analyses to assess what factors influence the choice of directional construction. Variables pertaining to the lexical verb and the directional morphemes are found to be the most important factors affecting the variation in the directional system. Sociolinguistic factors are shown to play a less significant role. The findings have implications for the grammaticalization of directional elements, as well as the typology of directional constructions in Oceanic languages.

show abstract

A Cross-Sectional Age Group Study of Coarticulatory Resistance: The Case of Late-Acquired Voiceless Fricatives in English

Howson

Redford

2022

J Speech Lang Hear Res

View full text Add to dashboard Cite

Purpose: As a class, fricatives are more “resistant” to consonant–vowel coarticulation than other English sounds. This study investigates the relative coarticulatory resistance of /θ, s, ʃ/ in child and adult speech to better understand the acquisition of individuated speech sounds. Method: Ten 5-year-old children, seven 8-year-old children, and nine college-age adults produced [əFV] sequences in carrier phrases, where F was /θ/, /s/, or /ʃ/ and V was /æ/, /i/, or /u/. In Experiment 1, coarticulation was perceptually indexed: 65 adults predicted the target stressed vowel based on forward-gated audiovisual speech samples for a subset of four speakers from each age group. In Experiment 2, dynamic spectral measures of the /əFV/ sequences were analyzed using smoothing spline analysis of variance to again test for vowel effects on fricative articulation across age groups. Results: The perceptual results indicated that fricatives blocked vowel–vowel coarticulation across speaker age groups. Contrary to expectation, vowels were most accurately predicted when F was /s/ and not when it was /ʃ/ or /θ/ across age groups. Acoustic results indicated the expected biomechanically motivated /ʃ/ > /s/ > /θ/ coarticulatory resistance hierarchy in adults' speech. By contrast, /ʃ/ > /s/ were similarly influenced by context in 8-year-olds' speech, and the results from 5-year-olds' speech suggested an influence of order of acquisition in that /θ/ was surprisingly resistant to coarticulation. Conclusion: The study results are taken to suggest that a temporal constraint on fricative articulation interacts with biomechanical constraints during development to influence patterns of coarticulation in school-age children's speech.

show abstract

Leveraging the temporal dynamics of anticipatory vowel-to-vowel coarticulation in linguistic prediction: A statistical modeling approach

Cited by 3 publications

References 70 publications

Advancements of Phonetics in the 21st century: Quantitative data analysis

Advancements of Phonetics in the 21st century: Quantitative data analysis

Directional constructions in Matukar Panau

A Cross-Sectional Age Group Study of Coarticulatory Resistance: The Case of Late-Acquired Voiceless Fricatives in English

Contact Info

Product

Resources

About