Variability and Central Tendencies in Speech Production

Whalen, D. H.; Chen, Weirong

doi:10.3389/fcomm.2019.00049

Cited by 9 publications

(4 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These distributions were characterized by their location (centroid) and spread parameters, the latter in the form of a 2 × 2 covariance matrix, which accounts for both spread and correlation between F1 and F2 values. This level of description amounts to assuming that formant values within each vowel category are distributed as two-dimensional Gaussians, a fairly common assumption in the literature ( Whalen & Chen, 2019 ). We then computed the MD between each token of a given vowel category produced by the mobile speakers and the corresponding reference (sedentary) distribution using formula B1 in Appendix B .…”

Section: Methodsmentioning

confidence: 99%

Using Mahalanobis Distances to Investigate Second Dialect Acquisition: A Study on Quebec French

2022

View full text Add to dashboard Cite

Second dialect acquisition (SDA) can be defined as the process through which geographically mobile individuals adapt to new dialect features of their first language. Two common methodological approaches in SDA studies could lead to underestimating the phonetic changes that mobile speakers may experience: only large phonetic differences between dialects are considered, and external sources are used to infer what should have been the speakers’ original dialect. By contrast, in this study, we carry out a longitudinal analysis to empirically assess the speakers’ baseline and shift away from it with no priors as to which features should change or not. Furthermore, we focus on Quebec French, a variety with a relatively crowded vowel space. Using Mahalanobis distances, we measure how acoustic characteristics of vowels produced by 15 mobile speakers change relative to those of a control group of 8 sedentary speakers, with the mobile participants recorded right after they moved to Quebec City, then a year later. Overall, the results show a reduction of Mahalanobis distances over time, indicating convergence toward the control system. Convergence also tends to be greater in denser areas of the vowel space. These results suggest that phonetic changes during SDA could be finer than previously thought. This study calls for the use of methodological approaches that can reveal such trends, and contributes to uncovering the extent of phonetic flexibility during adulthood.

show abstract

Section: Methodsmentioning

confidence: 99%

Using Mahalanobis Distances to Investigate Second Dialect Acquisition: A Study on Quebec French

2022

View full text Add to dashboard Cite

show abstract

“…Employing an adequate number of Gaussians, varying their means, covariance, and weights makes it possible to approximate any continuous density to a random accuracy. It is the most frequently used distribution system for modeling speaker and gender features in person and GR systems because speech features are usually assumed to be normally distributed 51 . The ease of learning ability, boot strapping from flat data faster computation, and compatibility with frame‐level features makes it a choice modeling tool.…”

Section: Methodsmentioning

confidence: 99%

“…It is the most frequently used distribution system for modeling speaker and gender features in person and GR systems because speech features are usually assumed to be normally distributed. 51 The ease of learning ability, boot strapping from flat data faster computation, and compatibility with frame-level features makes it a choice modeling tool. GMM has also been proven to be very efficient and effective in text-independent classification system where the prior knowledge of what the speaker will say next is not known.…”

mentioning

confidence: 99%

Laughter signature, a new approach to gender recognition

Folorunso

Popoola

Asaolu

2020

Engineering Reports

View full text Add to dashboard Cite

Gender Recognition (GR) is the process of identifying gender difference by extracting and evaluating features from data existing as images, video, audio, text, and other signals. The task of GR has been achieved using face, keystroke, gait, and speech features. "Laughter" with its intrinsic usefulness for this task is yet to be explored. Laughter is an important paralinguistic feature in human communication. It can change the meaning of speech when triggered by some form of arousal or amusement. Although, laughter can be "acted," but the human natural laughter is a spontaneous reflex response, which reasonably embeds some characteristic peculiarities of the individuals. The human brain has capacity to make this distinction. In this study, spontaneous laughter bouts of 123 volunteers (41 females and 82 males) were recorded. The Dynamic-Average of the Mel frequency Cepstral Coefficient (DA-MFCC) were generated and trained using two conventional and effective machine learning algorithms that have been employed in gender identification. These algorithms are Gaussian Mixture Model (GMM) and Support Vector Machine (SVM) classifiers. Overall accuracies of 87.65% and 86.91% were obtained with GMM and SVM respectively. Therefore, indicating the possibility of using laughter characteristics as signatures for distinguishing between male and female genders. For both classifiers, the use of DA-MFCC reasonably reduced training time. Some of the potential areas of applications of GR include security, health care, marketing, human machine interaction toward enhanced emotion recognition, automatic speaker recognition and forensics.

show abstract

“…Daria D'Alessandro, Cécile Fougeron CNRS, Grenoble, France University of Sorbonne Nouvelle, Paris, France Premising that it is planned, anticipatory Vowel-to-Vowel coarticulation provides a cue on the size of the encoded speech units, since it reflects the coordination of elements that are planned together in the same unit (e.g. Whalen & Chen, 2019). Several studies have shown that coarticulation varies, e.g., according to speakers (Zellou, 2017), age (D'Alessandro & Fougeron, 2021), rate (Matthies et al, 2001), style (Scarborough & Zellou, 2013).…”

Section: Variability In Anticipatory V-to-v Coarticulation In Frenchmentioning

confidence: 99%

Abstracts of the 8th International Conference on Speech Motor Control Groningen, August 2022

SSTP

2022

SSTP

View full text Add to dashboard Cite

This Supplement of ‘Tijdschrift voor Stem-, Spraak- en Taalpathologie’ (Journal of Voice, Speech and Language Pathology) contains the abstracts of the eighth edition of the International Conference on Speech Motor Control, held in Groningen, The Netherlands, August 24 - 27, 2022. With this eighth conference, a well-established Nijmegen (5 editions) - Groningen (6th & 7th edition) tradition continues. This conference, like the ones before, highlights new trends and state-of-the-art approaches in theoretical and applied research in the area of normal and disordered speech motor control. The five years since the previous conference in 2017 have yielded not only further insights in genetic, neural, physiological and developmental aspects of speech production, stuttering and other speech motor conditions, but have also advanced theoretical modelling. Combined with ongoing studies of populations that are increasingly better characterized genetically and neurobiologically, this quantitative boost of interdisciplinary results is now leading to a qualitative turning point in which large data sets are analyzed with powerful artificial intelligence and machine learning algorithms. The implementation of theories into computational models allows for the explicit testing of multifactorial interactions, thereby going beyond the more traditional single-factor experiments. Machine learning and the data sharing required to make this feasible are a special topic of the 2022 conference.

show abstract

Variability and Central Tendencies in Speech Production

Cited by 9 publications

References 37 publications

Using Mahalanobis Distances to Investigate Second Dialect Acquisition: A Study on Quebec French

Using Mahalanobis Distances to Investigate Second Dialect Acquisition: A Study on Quebec French

Laughter signature, a new approach to gender recognition

Abstracts of the 8th International Conference on Speech Motor Control Groningen, August 2022

Contact Info

Product

Resources

About